DE60025753T2 - SERVO CONTROL DEVICE WHICH USES THE ABSOLUTE VALUES OF INPUT SIGNALS - Google Patents

Abstract

A servo control system comprising the combination of: a signal source provides an error rate function f(y) corresponding to a displacement y , the displacement y characteristic of a system to be controlled; a sampling circuit for sampling the value of f(y) at a sampling frequency of X f to produce a function f(y) j ; a map responsive to f(y) j to produce a positive value and a negative value ; a first estimator responsive to to provide an estimated value y 1, j ; a second estimator responsive to to provide an estimated value y 2, j ; and nonlinear logic responsive to y 1, j and y 2, j to generate an estimate y k of the displacement.

Description

BACKGROUND THE INVENTION

AREA OF INVENTION

The The present invention relates to control systems and methods, and more particularly Control systems and methods that are useful when a system characteristic, the for Control purposes, values with the same sign and the same size for both positive as well as negative changes a variable that is set to control the system, can accept. While It is believed that aspects of this invention are in a very general way to many different control systems and procedures Aspects of the invention find their immediate applications on the position control of detectors used in data acquisition systems (Data acquisition systems) can be used. Specific embodiments of the invention are useful in servo-mechanical control systems, such as magnetic disk drives high-density, one of which data through accurate positioning a magnetic read element adjacent to a set of predefined ones Data storage locations are read.

2. DISCUSSION OF THE RELATED STATE

The related situation is illustrated by reference to several simple tax systems. A common task assigned to control systems is to maintain the relative positions of one object relative to another object, whereby both objects can move in an unpredictable manner. A useful system for viewing is an optical disc player which is schematically shown in FIG 1 and that is of the type that laser light from a laser 1 on a data storage surface 2 an optical disk to read data from the disk. Typically, the laser light diverges and becomes on the data storage surface 2 through an objective lens 3 focused. During operation, the disc (the disc) may bend or vibrate, causing the distance between the data storage surface 2 and the focal point of the objective lens 3 changes enough to measurably degrade the focus of the laser light on the surface of the optical disk being read. To prevent signal variations and degradation, the disc player adjusts the position of the lens to the distance between the lens 3 and the data storage surface 2 Here and throughout the discussion of the background and the invention, the term "nominal" has its usual meaning as satisfactory or on schedule.

It is typically not practical to measure the distance between the lens 3 and the data storage system so that disk players indirectly observe this distance. For example, light reflected from the data storage surface may be through a beam splitter 4 go through the lens 5 collected and refocused and to an optical detector 6 be directed. The optical detector 6 is divided into four quadrants, as in 2 shown. The system is constructed so that when the distance between the lens 3 and the data storage surface 2 equal to the nominal "focus" distance that is on the detector 6 incident light is focused (focused) and has an intensity distribution that changes symmetrically on the surface of the detector. Such a symmetric in focus state is in 3 shown. Each quadrant of the detector provides a separate output voltage V _A , V _B , V _C , V _D such that the symmetrical state of the 3 is associated with a value of effectively zero the observable magnitude (V _A + V _D ) - (V _B + V _C ).

In the laser light, which is used to read the data storage surface 2 is used, an asymmetry is introduced, leaving too short a distance between the lens 3 and the data storage surface 2 an asymmetric non-in-focus pattern on the detector 6 generated like the one in 2 is shown. This asymmetry is characterized by high intensity light on quadrants B and C of the detector 6 and by low-intensity light on quadrants A and D. Too long a distance between the lens 3 and the data storage surface 2 creates the non-in-focus pattern that is in 4 represented by high-intensity light on the quadrants A and D of the detector 6 and is characterized by low intensity light on quadrants B and C. The amount of asymmetry varies with the amount by which the distance between the lens 3 and the data storage surface 2 As a result, the quantity (V _A + V _D ) - (V _B + V _C ) may be a useful variable to be observed for the position of the lens 3 to control. For example, a linear control system may be provided to adjust the lateral position of the lens 3 to set a good control of the focus of the optical disk system of 1 using the magnitude (V _A + V _D ) - (V _B + V _C ) as an input.

In the system of 1 The magnitude (V _A + V _D ) - (V _B + V _C ) provides a good observable variable for controlling the position of the lens 3 (the controlled variable) and thus the focus of the system. (V _A + V _D ) - (V _B + V _C ) provides both a magnitude indicating the extent of the required correction and the sign of the required correction. The sign of the required correction indicates whether the lens 3 too close or too far away from the data storage interface 2 and thus, in which lateral direction the lens 3 must be moved to optimize the focus. (V _A + V _D ) - (V _B + V _C ) can be used to control the system of 1 because (1) the laser light is conveniently made asymmetric within the system, and (2) the detector is formed of four independent quadrant detectors. In other words, the optical system is specially designed to easily control the position of the lens 3 to allow.

While it is possible the system of 1 using the asymmetric beam in conjunction with a quadrant detector, this is not a completely desirable situation. It is easier to construct optical systems for light that changes uniformly than it is to construct optical systems of similar quality for asymmetrically varying beams. This is especially true when different colors of light could be used in the optical system.

Changes in the system of 1 that do not intentionally introduce some sort of asymmetry into this beam of light could have a detector like the one in 5 is shown using an outer element 7 which generates an intensity-dependent voltage output of V _T , and an inner element 9 , which generates an intensity-dependent voltage output V _C , includes. Using such a system, the magnitude V _C / (V _C + V _T ), which is schematically represented as a function of the distance S of the lens to the memory surface from the nominal in focus position S ₀ , provides a measure of the focus It is possible to use this observable variable to adjust the focus of the system, but it is difficult, because the magnitude V _C / (V _C + V _T ) indicates a magnitude of a correction to be made, but not For each observed change in magnitude V _C / (V _C + V _T ) from the peak, in the focus value, the lens might require a shift in either a positive or a negative direction. It is thus difficult to change the system 1 directly using only the size V _C / (V _C + V _T ) to control.

Systems like those in 6 may be referred to as absolute value systems because the observed variable provides the absolute value of a correction to be made, but not the sign or direction for correction. For such absolute value systems, it is known to introduce motion in a known direction, such as oscillation, to the controlled variable. This technique is known as dithering or dithering. Using a jitter, there will be a regular oscillation to the position of the lens 3 which introduces a regular change in the observed V _C / (V _C + V _T ). By comparing the phase of the oscillations in V _C / (V _C + V _T ) with the phase of the oscillation in the position of the lens 3 For example, the control system may identify the direction of correction to be made for the lens position. Consequently, a tremor, which in the 5 and 6 shown mechanism is used to the optical disk player of 1 to control.

Dithering has a variety of disadvantages. It is complex, requiring the introduction of a detectable amount of movement between objects. Furthermore, the jitter itself is a source of noise and is thus generally undesirable. Consequently, it is more common to construct a system such that it has an observable variable that provides both a magnitude and a direction or sign for correction, such as a system such as the one described in U.S. Pat 5 and 6 is shown to use.

The above example relates to the control of the optics of an optical disc player and illustrates the difficulty of attempting to control such a system with an absolute value using an observable variable. In this sense, an observable variable with an absolute value is one that varies with the variable to be controlled, but only in size, as with the graph of the 6 A similar problem can be illustrated with reference to magnetic disk storage systems and this discussion is now directed to this technology. This background discussion illustrates aspects of the problem that will address aspects of the present invention, as well as aspects of an environment in which certain preferred aspects of the present invention are implemented.

Data storage systems with a magnetic disk drive provide a long-term data storage high volume ready, comparatively fast and relatively economical is at least as measured on a per-bit basis. For example Magnetic disk storage is faster than current optical storage options and is relatively cheaper as present on a flash memory-based Storage devices. Today's industry relies on magnetic bar drives for one Long-term data storage in different types of computer systems and in certain electronic consumer applications, such as the Video recording and playback, and both types of uses are growing continue on. Research in magnetic disk drive systems continues and it is expected that the performance of such Systems continues to improve.

Rotating storage devices and especially disk drives store data on or several sides of a rotating medium, often as plates or discs is called. For the case of a conventional one Disk drive data by generating a magnetic Modulation within a magnetic material on a Data storage surface the plate is stacked stored. Data becomes thereby read back after that this modulation is detected with a read head. typically, Data is sent to the disk using a write element and data is written from the disk using a Read element, with both the write and read elements provided as physically separate elements on a single head become. By recording data in the form of magnetic Signals on the rotating disk can both store data will be recovered as well as afterwards even after long periods of time. Data can be stored in a plurality of radially offset, tangentially extending ones Tracks are organized, with data on the tracks in general into a multitude of data blocks be stored organized. To get data from a specific block of data to read or write data to a specific data block, The typical disk drive positions the read and write head over the disk drive Track containing the destination block of data with a function which is known as search mode (Seek operation). The reading and Write head of the disk drive reads or writes the data the storage surface as desired. Data read and write operations, search operations and other operations, such as the use of Gray Codes for identifying lane locations are described in U.S. Pat. U.S. Patent No. 5,523,902, U.S. Patent No. 5796543 and U.S. Pat. patent No. 5847894.

7 schematically illustrates certain aspects of a storage surface 10 and a disk drive. In general, the plate includes 8th a central opening 12 through which the spindle of the disk drive passes and with which the disk and its storage surface are rotated. An area 14 gets on the plate around the central opening 12 provided around to clamp the plate to the spindle or keep it somehow different; this area 14 is essentially unusable for storage. The storage surface 10 extends radially away from the nip area 14 and may terminate in a peripheral tape of the disk, not shown, and also preferably not used for data storage. In general, the storage surfaces of disk drives may be considered substantially uniform or the storage surface could be divided into multiple zones. If the storage surface 10 When treated as substantially uniform, relatively little is done in the disk drive to account for differences between radially offset data storage locations, such as differing rotational speeds, and the associated differences in areal density of stored data on the disk surface. Another strategy shares the storage surface 10 into a number of radially extending bands known as zones, such as the zones 16 . 18 and 20 on the storage surface 10 are indicated on. The different storage locations within the different zones 16 . 18 and 20 are treated similarly while the memory locations in different zones may be treated differently, for example by using different clock rates to read or write different signals, or by using different densities of the servo information.

The read and write head 22 is a small arrangement on the end of an arm or a transducer assembly 24 is provided, the head 22 over the storage surface 10 emotional. The transducer assembly can head 22 by rotation, by translation, or by a combination of rotations and translations. For example, many current drives make larger movements by rotating the transducer assembly about pivoting on the end of the transducer assembly opposite that of the head 22 ready. Additional adjustments can be achieved using fine translations that could be achieved, for example, using piezoelectric elements. In general, the mechanical rotational and translational movements of the head 22 under servo control using, for example, voice coil motors or other compact systems with fast response preferably realized. The read and write head 22 of the transducer assembly is typically not fixedly attached to the transducer assembly. Instead, the read and write head is preferably mounted on a slider which is coupled to the transducer assembly via a flexible structure. Typically, the slider is designed to "fly" on an air bearing over the data storage surface created between the slider's shaped sub-slide and the disk.

8th shows in its details aspects of a presently preferred read and write head design. The illustrated read and write head 22 is on one end of a slider 26 attached, in turn, to the transducer assembly (in 8th not shown) is attached. A magnetoresistive read element 28 is formed as a thin film element near or on the end surface of the slider, and then becomes a writing element 30 an inductive or other type partially over the read element 28 intended. A protective coating 32 covers the read and write head 22 from. As shown, it is typical that the writing element 30 much larger (sometimes 160% or more) than the read element 28 is. In addition, the read element is typically offset to one side relative to the write element. This configuration is characteristic of the illustrated types of elements and causes the read and write heads to have different preferred positions relative to a track or other data storage structure for respective read and write operations.

A read and write head 22 belonging to a storage surface is accurately positioned with respect to data storage locations along a track through the use of servo control mechanisms within the disk drive which work in conjunction with location server information stored on the storage surface of the disk drive. Various servo techniques have historically been used for magnetic disk drives, the industry currently favoring the use of servo information contained on each data storage surface on the disks within the disk drive. When reliably performing a track seek operation, the disk drive uses the read element 28 of the head to detect servo position information used by control circuitry to position the transducer assembly and head above the target track. The servo position information identifies the position of each track and provides at least relative identification for each of the tracks on the disk drive.

Positionssteuerungs- or servo information will often extend radially within itself Sector servo wedges stored, which are described in more detail in the above described patents, and on the data storage surface of the Plate during the original one Preparation of the disk storage device are placed accurately. The position and other servo information can with a servo writer like the one disclosed in U.S. Pat. Patent No. 4920442 is described, or in agreement written by the methods described above. Servo writers are used in a factory initialization process, to position and other servo information on the memory surfaces of the Plates, along with other information, to write to the storage surface prepare for use. The servo writer, typically accurate position information used by a laser positioning mechanism is most often placed on servo information each lane along predefined radial spokes indicating the beginning from each sector on the disk.

7 shows two possible organizations of servo information on the storage surface of a disk 8th the one discussed above with the full radial wedges 34 about the usable radial extent of the data storage surface 10 extend, and another, at the Teilservokeile 36 be provided with different seal in different zones of a storage surface. The first method may have bursts of the order of 100-200 servo bursts positioned at regular angular intervals on the memory surface of a 3.5 "disk. Different design and operating parameters can greatly change these characteristics. In the second method, there are an increasing number of servo wedges in each of the zones as they go away from the center of the plate. In general, only one of the two schematically illustrated methods is used on a plate.

Regardless of whether zones on the data storage surface are differentiated, the servo wedges may include a substantial amount of information useful for positioning the head and reading and writing data to the disk. An illustration of the information that may be included in the servo wedge is provided, for example, in US Pat. No. 5,796,543 and is incorporated herein by reference 9A & 9B reproduced, each representing writing and reading operations. These figures show a portion of a servo wedge 40 extending over four data tracks Tr 0, Tr 1, Tr 2 and Tr 3. The wedge 40 is from a servo preamble 42 and servo position information 44 formed and follow the wedge one or more data blocks 46 in each of the tracks. A head 22 including both read and write elements is shown to indicate the progress of data (to the left) through the head and the preferred position of the head relative to the centerline of the track during the write operation ( 9A ) and the reading process ( 9B ). As shown, the read element is typically held a predetermined distance away from the centerline during a write operation. Therefore, the preferred tracking position is sometimes adjacent to the centerline of the track.

The servo preamble 42 provides information used to set the read channel electronics to read and process the position servo information. The servo position section 44 of the servo wedge provides the actual position data that passes through the read element 28 read and positioning the head 22 should be used. The illustrated servo preamble 42 starts with a pre-burst gap 48 in which no transitions are recorded, followed by a field 50 for Automatic Gain Control (AGC), which could potentially include a regular pattern of transitions (eg, a positive 3T pattern followed by a negative 3T pattern) used to adjust the gain of the read channel electronics. The servo preamble next contains a synchronization pattern 52 for setting the clock in the read channel electronics when the servo position information is read, this one servo address mark 54 that the read channel electronics indicates that the subsequent information, unlike data, will be servo position information. Next, the servo preamble 42 an index field 56 which provides position information within the track, ie, whether the servo wedge is the one referred to as the first servo wedge on the track.

After the servo preamble 42 is the servo position information 44 including rough position information 58 and fine position information 60 - 66 , The rough position information 58 For example, Gray may include codes that numerically indicate each of the tracks on the storage surface. In general, a gap separates the rough position information 58 , the finer track positioning information provided by servo bursts 60 - 66 provided. The checkerboard pattern 60 - 66 offset (offset) servo bursts A, B, C, D of recorded information have been written to have accurate and desired positions with respect to the centerlines of different tracks within a given grouping of tracks. This allows the read element to generate a control signal related to the linear offset with respect to the desired position relative to a track, such as the track centerline, which control signal can be used to adjust the position of the head with respect to the track.

The shown checkerboard pattern, the A, B, C, D servo bursts is made by a servo writer using multiple Write and delete runs during the Manufactured so that each of the servo wedges shown Pattern of four rectangular Servo bursts, repeated at desired radial and tangential positions. The servo bursts A, B, C, D could be internal for example, consist of a repeating 3T pattern, where the servo bursts are surrounded with no recorded transitions. In the normal Operation of the disk drive will be the limits of servo bursts in track seek and tracking operations to detect a position error signal Periodically generate (Position Error Signal, PES) that uses can be to the position of a head in relation to a data track adjust. Between the servo bursts are several (typically 3-5) Data blocks along the track saved. The servo control mechanism works in cooperation with the buried servo information to the head exactly at one desired Position relative to the track to place when the servo burst passes under the head. No additional positioning information is available until the next Servokeil passes the head. As a result, the servo control mechanism tries the head at a fixed position with respect to the track position, by the youngest Servo Burst is identified to hold. It is possible that the head or the plate as a result of mechanical stops, vibrations, move thermal variants or other disturbances in the system, before the next Servo burst is achieved.

In addition to the track identification information within the servo wedge, memory surfaces are sometimes provided with additional information to indicate when the desired track and sector have been located in a seek mode. An ID header block may optionally be provided between the servo burst and the first data block of a sector. The ID header mainly includes an identification for the track and the following sector. Aspects of the use of synchronization patterns and headers are described, for example, in US Pat. No. 5,541,783 and US Pat. No. 5,796,534. While header information at the start of sectors in many systems Other techniques for identifying tracks that do not use headers are known. For example, ID and header information may be incorporated into the servo bursts, as in the article by Finch, et al., "Headerless Disk Formatting: Making Room for More Data", Data Storage (April 1997), p. 51 54 or servo information may be cross reference information stored in a corresponding table in the memory, as described in IBM Storage Publication Hetzler, "No-ID Sector Format", dated January 8, 1996.

A servo wedge 40 (or an ID header block, if used), several blocks of data (typically 3-5) may be stored along a track, as in FIG 10 shown. Each data block 70 . 72 . 74 includes a data synchronization pattern 76 . 78 . 80 , which is positioned next to the data storage area of the block. Typically follows a block of data 70 . 72 . 74 an EEC block 82 . 84 which stores error identification and correction codes for the previous frame. The data storage area of each data block is typically of a size sufficient to store data signals representing 512 bytes of data.

The data synchronization pattern includes synchronization information that can be extracted to establish a sampling frequency and a phase for recovering data from the data storage area. The conventional synchronization pattern 76 . 78 . 80 is determined by the write element of a header in an operation where the associated data blocks 70 . 72 . 74 written, written. The clock rate used to write the sync pattern is also used to write the subsequent data blocks. During a subsequent read operation, the read element of a head passes over the synchronization pattern and detects a pattern of transitions (eg, a 2T or 3T pattern), from which a clock for reading the subsequent data blocks is derived. Conventionally, it is preferable that the synchronization pattern is substantially uniform in the radial direction, changing only in the tangential direction for a read element positioned in a desired manner with respect to the track. Typically, the disk control logic and actual disk rotation rate determines the data rate written for the data synchronization pattern and the data storage area that follows. As a result, the actual data rate may change from block to block and sector to sector, and consequently the amount of space occupied by the stored data may change. To accommodate these changes, there is typically a gap (an inter-block gap) or a data patch 86 . 88 after each data block to ensure that there is sufficient physical separation between successive blocks along a track to allow data blocks to be written without overwriting a subsequent block header or backward servo burst.

It gives a tension in the design of data storage systems between an increase the track density, which typically requires denser servo die patterns, and the loss of the storage area as a result of the provision from servo information with elevated Densities. It is desirable additional Servo information without the area of the data storage surface, which for the actual Storage of data is intended to reduce. With others Words it is desirable the storage density without increase overhead, required for accurate saving and reading out of Information, increase.

11 Figure 12 shows the differences between accurate position control during a write operation and worse position control during a write operation, as might occur, for example, for a data block shifted from a sector servo when the disk is subjected to vibration. The figure is a data block within a well written track 90 where the written data is symmetrically arranged around the centerline of the track. The writing width of the track (TW) is arranged symmetrically within the distance allowed for the track TP. For a well-written track 90 As illustrated, a read head centered on a track detects a read area 92 which is positioned within the well-stored data on the track, between the reading area 92 and the edges of the described track area. The track arrangement could, for example, have a track pitch TP of 1.0 μm, a write width TW of 0.8 μm, and a read width 92 of 0.6 μm.

Unlike the well written track 90 , becomes the badly described track 94 poor position control and mechanical failure. The path of the poorly written track changes within the track pitch in an irregular manner. It is possible that such an irregular write path is sufficiently misaligned that the read width is partially away from the described area, as in FIG 94 displayed. Such misalignment reduces the Quality of the data read out and can lead to read errors. To avoid this problem, conventionally, the size of the writing width is increased relative to the reading width. Another mistake with 96 in 11 is indicated, occurs when a write head is misaligned to a sufficient extent to overwrite the described area of an adjacent track. Such an error is extremely undesirable and is avoided by increasing the track pitch.

Each of the potential mistakes in 11 are conventionally addressed by increasing the spacing between structures within the track. Therefore, improvements in head positioning accuracy can be significant factors in improving the density of tracks and written information.

The foregoing discussion has listed aspects of the servo and other informational structures provided on the magnetic disk drives and how this information is used. With this background presented, the discussion now reverts to control systems for magnetic disk drives and how different observable variables are used in such systems. Typically, Position Error Signals (PES), which are monitored to control the position of the read and write heads with respect to the track position, are derived as dedicated servo patterns. In particular, fine position control is performed by deriving PES from the A, B, C, D servo burst patterns included in the 9A & 9B are shown carried out. The position error signals detected during translation of the read head via the A, B, C, D servo burst patterns change linearly around zero, thus providing both magnitude and direction of corrections for the position of the head should be performed. Here, the detected position error signals are the observable variables that can be used to control the position of the head relative to the data track. As described above with reference to the 10 and 11 however, the control in accordance with the conventional servo information is not fully sufficient to meet the current and assumed requirements for magnetic memory systems. Alternative control strategies are desirable.

The U.S. Patent No. 5233487 by Christensen et. al. refers to Magnetic disk drives and the control of the head position in relation on a track using an absolute value observable variables for the tax system. The Christensen patent describes a disk drive, which periodically interrupts data read and write operations, to calibrate the optimal head offset with respect to a Data track for Perform data write operations. The calibration is done by a read head on the Trace through a range of intended misalignments in With respect to the track is translationally moved. Data will be sent to the deliberately read many aligned positions and the errors which occur when reading from the track are considered a function get the misaligned position. Error rates are going through Capture blocks of data and then by applying the standard data block error correction method using the ECC circuitry of the disk drive in terms of ECC data with those stored on the disk Data are encoded. The disk drive detects at which track position and thus at what level of deliberate misalignment the observed error rate determines certain target error rates with a known one Association with known track positions exceeds. This allows that the characteristics of the working disk drive with known ones Standards are compared.

It should be noted that the target error rates (in 2 of the Christensen patent) are absolute value functions. As such, the target error rates derived in the system of the Christensen patent do not indicate the direction in which the head is misaligned with respect to the data track. The sign of misalignment is known by the control system because of the intended positioning of the read head, much in the same way that dithering can be used to create a known direction for correction within an absolute-observable control system.

The track positions detected using the method of the Christensen patent can be used to calculate a track pitch position, to recalibrate the relative motion obtained by applying a controlled offset voltage, and to provide an optimal offset for reading data from and to Write data to the track set up. In this way, the Christensen patent provides a servo control function using information other than the servo wedge and the servo burst information included in the 9A and 9B is shown off. In particular, the Christensen patent executes servo control using error information derived from constraints associated with blocks of data stored on the disk. The procedure of However, Christensen's patent does not allow derivation of position error signals on a continuous basis or while data is being extracted from the magnetic disk.

The Patent by Christensen leads Servo functions using an observable variable, a bit error rate derived from ECC operations for data blocks that only provides absolute value information regarding the position error. On the other hand, the practice of the method of the patent requires Christensen used a head-to-track displacement, many of them Has problems with the shaking and therefore not used can be when data is collected. It is desirable to provide a method and system capable of is to use an absolute value information to a control function perform.

WHERE 99/36907 describes a disk drive system with a servo circuit, which is an error signal for one predicted position (PES) as an approximation of the actual PES generates and then a head position by a summation of the actual Controls PES with the predicted PES. The actual PES is nominally called not linear over characterized by a portion of the width of a selected track.

The EP 0 298 475 describes a magnetic head positioning feedback control system that uses mathematical algorithms to compensate for tracking error due to noise or disk destruction. The system comprises a prediction unit (predictor) for generating a predicted drive signal in accordance with previous head drive signals stored in a memory.

The EP 0 331 189 describes a position detecting method in which the position of a moving body based on two-phase periodic waves having a phase difference smaller than one period is detected. The method generates a highly estimated position and a finely estimated position and corrects an error due to hysteresis at the rough estimated position starting point by using the relationship between the roughly estimated and finely estimated positions.

The US 4,217,612 describes a servo system for a magnetic disk drive in which recordings on the magnetic disk are carried out with data signals and interleaved servo signals. A first position error signal is obtained from the servo signals, and a second position error signal is provided by a track increment generator. The first and second error signals are summed to produce a combined position error signal.

The US 4,166,282 describes a method and apparatus for positioning a bi-gap converter relative to a selected track on a magnetic disk. "Servo bits" providing head positioning information and "working bits" may both be recorded along a given track. Once the bi-gap head is moved along a given track, one gap is aligned with the servo bits and the other with the work bits. The phase of the servo signal values indicates the degree of head alignment.

The US 5,966,264 describes a two-frequency servo position error detection pattern. The servo pattern has bursts that are constructed of two frequencies that are orthogonal - ie, 90 ° out of phase - and that are arranged to capture both frequencies simultaneously to form one of three phase signals and to detect a position error detection (PES). Signal to provide. The three-phase PES signal is said to be sufficient to provide substantially the same information provided by a conventional quad-burst servo pattern and therefore only three types of servo bursts A, B and C for the present invention underlying tri-burst servo patterns are needed, eliminating something that would otherwise be the D-pattern bursts for a quadrature servo pattern, reducing the disk surface area required for the tri-burst servo pattern by 25%. is reduced.

EP 0 690 442 describes a tracking servo control system in a magnetic media storage device. which derives head position information from one or more specially structured servo tracks. The servo patterns are formed of magnetic transitions recorded in more than one azimuthal orientation in a servo track so that the timing of the servo position signal pulses read from reading the servo pattern at any point on the pattern continuously changes as the head transitions the width of the servo track is moved. The timing of pulses read by the servo read head is decoded to provide a velocity invariant position signal which is used by the servo system to sweep the data heads over the desired data pulse to position on the storage medium. The servo pattern is formed of a repeating cyclic sequence containing azimuthal orientations with two different transitions. The pattern may comprise straight transitions that are substantially perpendicular to the length of the track, alternating with azimuthally inclined or oblique transitions. That is, the azimuthally inclined transitions extend across the width of a track at an angle to the direction of head movement.

The US 4,048,660 describes a servo signal pattern that includes first and second sets of servo signals that are transversely aligned to track localization lines on a magnetic disk. The servo signal groups are spaced longitudinally along the lines. A servo signal in each of the groups has a transverse extent to the lines about the distance from adjacent ones of the locating lines and having different lengths or lengths along the length of the lines. Servo signals having shorter lengths are preferably longitudinally adjacent signals of longer lengths in the respective groups. Certain servo signals abut at least one of the localization lines, while other servo signals are centered on other localization lines. The servo blocks may have different frequencies or correlation patterns.

SUMMARY THE PREFERRED EMBODIMENTS

It It is an object of the present invention to provide a method of controlling at least one aspect of a system based on an observable one Provide variables that are either positive for both negative variations increase by a nominal or target value or for both positive and negative changes falls around the nominal or target value. Such an observable one Variable has the property that a value of the observable Variables on two or more values of a variable that is controlled Aspect of the system depicted is such an observable one Variable is identified here as an absolute variable and that System as an absolute value system.

aspects of the present invention provide a servo control device and a method that can be combined to form an absolute value system based on an absolute value observable To control variables. An observed value of such an "absolute value" variable could be either a positive or negative displacement of the controlled variables of their nominal value. The possibility of a picture of the observed value on two different shifts or Offsets make it difficult to control the system. This is because the value of the variables observed is not the direction indicates in which the controlled variable should be set, closer to the controlled variable to move to their nominal value. Aspects of the present invention allow control of systems at least partially below Using an observable variable, the absolute value characteristics with respect to the controlled variable.

One preferred aspect of a system in accordance with the invention observes a value, the observable variable and the system make this value for two possible ones Shifts or dislocations corresponding to the observed Value off. Two estimation units are provided, with each appraiser being designed to match the current or future State of the system based on a set of input values, including the transfer, predict. The first estimator (the first estimator) takes one of the possible Shifts as an input. The second appraiser takes the other of the possible Shifts as an input. Both appraisers are used to values predict the observed variable after the system has passed through a time interval has gone. The absolute values of the observable Variables by the appraiser be predicted to be re-observed with one or more Values of the observable variables compared to determine which appraiser the correct assumption regarding the sign or the direction of the initial Represents displacement. Control functions are preferably under Use of the sign and the offset executed be identified as correct in this operation.

Further the wrong estimator is preferred to the condition of the right one Appraiser reset after the sign or the direction of the displacement predicted has been.

When applied to a position control system for an object, a preferred embodiment of the control system observes a value of a position error signal and maps that value to two potentially correct displacement values. Two estimators within the control system are initiated, one using one of the possible correct shifts and the other using the other shift (offset), and the estimators predict the future object position and the corresponding position error signal for each estimated future position. A new absolute value position error The detection signal is detected and compared with the absolute value of the two estimated position error signals. After sufficient system operation, the control system may make a selection as to whether one or the other of the estimators is correct. The correct shift, including its sign, can be identified as correct and used for future positioning operations, preferably until the sign of the object's displacement again becomes ambiguous.

The Control procedure, the system and the device find a special one advantageous application in the disk drive servo control systems, where the object undergoing positional control is is the head of the disk drive. In particular, the absolute value, which is observable and which is used in controlling the signal a signal derived from the data or the data signals that gets out of the storage surface read the disk inside the hard disk drive. To the Example could the observable absolute value is an error information characteristic represent the data or data signals recovered from the disk become. Alternatively could the observable position error signal in accordance with a boundary condition characteristic the data or data signals read from the disk be derived.

One Aspect of the present invention relates to a control system for Use in setting a controlled variable that represents a system to be controlled, wherein the controlled variable is represented within an observable variable and at least two values of the controlled variables a single value of the observable Correspond to variables. The control system includes a signal source, providing a first and second value of the observable variables adjusts the second value of the observable variable in time the first value of the observable variable. An illustration logic receives the first value of the observable variable and gives a first and a second possible Value of the controlled variables. Two appraisers are planned. A first Appraisers is capable of a future state of the system to be controlled, wherein the first estimator as an entrance the first possible Value of the controlled variable takes and a first output variable, the first predicted value of the observable variable represents, in response to the first possible value of the controlled Generates variables. A second appraiser is able to get one future Estimate the state of the system to be controlled, the second estimator as an entrance the second possible Value of the controlled variable takes and a second output variable, the second predicted value of the observable variable represents, in response to the second possible value of the controlled Generates variables. A determination logic determines which of the first and second predicted values of the observable variables, more specifically the second value of the observable variable.

SUMMARY THE DRAWINGS

These and other aspects of the present invention along with numerous obvious benefits related to use, and a practice of these aspects can be better understood by reference to understand the drawings that form part of the revelation.

The Drawings show:

1 Aspects of an optical disk playback system;

2 - 4 the use of an asymmetric beam in combination with a quadrant detector to produce an observable variable which is both magnitude and sign for controlling the position of an objective lens within the system of the present invention 1 provides;

5 & 6 the problems associated with the use of a symmetrical beam and a symmetric detector to generate a suitable observable variable for controlling the system of 1 ;

7 schematic aspects of a magnetic disk drive including a read and write head mounted on an arm and the data storage surface of a disk;

8th Aspects of read and write elements of a head;

9A & 9B information stored within a sector servo wedge and the relationship between a head, the servo information, and data tracks for read operations;

10 Aspects of the structure of data blocks within a track;

11 schematically the distance between tracks and a comparison between well-described and poorly described tracks;

12A & 12B Aspects of a method for identifying a correct correction to be applied in response to an observed position error signal;

13 schematically two different idealized error signals, each determined from data stored on the disk, in accordance with a constraint within the stored data. Both illustrated error signals exhibit a characteristic referred to herein as an absolute value relationship for positive and negative displacements (dislocations) about a nominal position;

14 empirical data that could correspond to an error signal derived from reading data from a data track with intended misalignments between the head and the data track. The vertical axis represents an error signal multiplied by 1,000, the horizontal axis represents distances in micro-inches measured from any origin with respect to a data track;

15 schematically illustrates the path of a read head to and around a nominal position within a target track;

16 a simplified block diagram of a servo control system in accordance with the invention;

17 a block diagram of the logic circuits within a servo control system, such as that used by a disk drive to estimate signals in accordance with the invention;

18 another position error signal with additional information related to the use of the position error signal graph as a non-linear map in a system partially in 17 is shown;

19 a diagrammatic plot showing the relationship between the different information rates and the timing of the nonlinear logic of the 17 is provided shows;

20 Aspects of a control system and a servomechanism consistent with the system described in the block diagram of 17 is shown;

21 a comparison of an actual position error signal with its estimation, as a function of time generated by the logic circuit shown in block diagram form in FIG 17 is shown;

22 as a function of time during and after a pulse disturbance, the position error signal with and without the estimated position error signal and control functions of one embodiment of the present invention;

23 as a function of time during and after a seek operation, the position error signal with and without the estimated position error signal and control functions of one embodiment of the present invention;

24 Aspects of a read channel circuit in accordance with a preferred embodiment of the present invention;

25 Aspects of data decoding within a decoder of the Viterbi or similar trellis type.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

The present invention provides a control method, system, and apparatus capable of controlling an absolute value system using an absolute value observable variable as an input variable. As used herein, one is in terms of the absolute value such variable, where the variable either rises or falls by a nominal or target value for positive as well as negative changes. A sufficiently large value, the variable observable in terms of the absolute value, indicates a displacement to be corrected but could represent a positive displacement or a negative displacement. Control functions based on the observable variables preferably take into account the lack of sign or the direction indication for the correction to be made.

A very short introduction is useful here a work framework to illustrate aspects of the present Invention to provide.

control systems Include models, either implicit or explicit, about how the system being controlled is responding to input data. For example, include most magnetic disk drive control systems an explicit model of the disk drive that predicts how the position of a read head in response to an application a certain voltage to an actuator changes. The disk drive control system Determines how to determine the head of the disk drive by putting under Using a model of the disk drive and the response the actuator is predicted, i. it is calculated how much tension should be applied to the actuator to the head to its desired position to move. The model uses the state of the disk drive, such as the current one Position of the head, as well as aspects of the history of the disk drive, and receives input data about the changes, to be performed on the disk drive. From this Information says the model within the control system of the disk drive the system behavior before.

These Discussion refers to appraisers who are the future ones or present State of a system based on a model of the current or future Predict or estimate system behavior. It is useful an appraiser so consider that it is based on a state of a system a model of the system, the history of the system and the inputs to the system System estimates or predicts, but changes from this framework are possible. In the technical field becomes more detailed in this discussion state estimator To the extent that the estimator has different characteristics of a system that together is exactly one system describe. Persons skilled in the art of the Control technology will recognize that the term "observer" ("observer") is sometimes interchangeable with the term estimator is used. The following description describes the control logic, the two slightly other appraisers used. For easier understanding It can be assumed that every appraiser has the same model of Systems, the same input data, the observed system behavior represent, and mostly the same earlier History of the system used. The two appraisers each comprise slightly different ones Assumptions about what happened and what happened in the system under control is happening. By comparing the estimates of system behavior, that through these two appraisers is shown with the actual observed behavior of the system, can determine the control logic, which of the two appraisers the more detailed description of the system has included. The tax system can then determine system characteristics based on the estimated sign and set shift information.

aspects The invention will now be illustrated by way of example the variable to be controlled the displacement of an object from whose nominal target position is and the observable variable represents a position error signal. Shifts can be positive or negative with respect to the nominal target position and that Position error signal has the same sign for positive and negative shifts. A sufficiently large observed Value of the position error signal shows a shift to be corrected on, could but a positive shift or a negative shift represent. In other words, the position error signal points as a function of the shift, an absolute value characteristic. The position error signal may be an asymmetric, non-linear May have relationship to the shift, or may have a simpler relationship to have. Thus, any observed position error signal corresponds to two different possible Shifts, where the two shifts have different signs and could have different sizes. Systems with even more complicated Pictures are possible and can in a similar Be processed, typically by using additional Appraisers.

In accordance with aspects of the present invention, this exemplary system may be controlled based on the observed position error signal using a knowledge of the mapping between the position error signal and the displacement, in conjunction with a model of the system. The position error signal is observed at a first time interval. The first position error signal could correspond to a positive shift + D1 or a negative shift -D2 of the object with respect to its nominal position, so there is ambiguity about which direction the object is to be moved in order to place the object at its nominal target position. The control logic initiates a pair of estimators, one assuming that the positive displacement + D1 is accurate, the other assuming that the negative displacement -D2 is accurate. Each two estimators predict the next position of the object at the next time interval. The estimators also predict values of the position error signal corresponding to the predicted shifts. At the next time interval, the position error signal observed for the absolute value is detected and then compared with the absolute value of the corresponding position error signals predicted by the estimators. The comparison of the absolute values of the position error signals predicted by the estimators with the observed position error signal preferably identifies the estimator with the correctly assumed sign or the correctly assumed direction of the displacement. Control operations are then performed using both the mapped magnitude and the estimated sign. Preferably, the system resets the state of the other estimator after the sign is estimated.

Consequently the control system observes an initial position error signal corresponding to a shift. The control logic keeps track of where the object would be if the initial one Shift was positive (estimator 1) and keeps track of where the object would be if the initial one Shift was negative (estimator 2). The appraisers say the displacement of the object under the different assumptions in advance of the sign of the initial shift and predict the position error signal which is for each of the different initial ones Sign assumptions would be observed after the system designed so that the object will be at a new shift would. The absolute value of the position error signal becomes later for the object observed and with the corresponding absolute values, the position error signal, that through the two appraisers predicted, compare. The state estimator with the position error signal which is closer to the measured absolute value position error signal is in absolute value, being considered the appraiser with the correct sign or the right initial one Shift identified. The comparison can be repeated at subsequent Time intervals performed until it is clear which of the two appraisers is the correct supposition about that Sign of the initial Displacement used. Thus, after a sufficient time or after a sufficient number of time intervals in one discrete system one of the appraisers predict a position error signal which is sufficiently different in the absolute value of the observed position error signal is the right appraiser chosen with confidence can be. This provides the control system with sufficient information to adjust the position of the object and the system based on to control a variable observable in terms of the absolute value.

A schematic representation of a process flow in accordance with the method just described is used as a reference in the 12A and 12B provided. An application of this control method to an optical disk system such as that incorporated in the 1 & 5 - 6 which is such as to provide an observable variable having a functional absolute value relationship with the variable to be controlled is readily apparent. The explanation of aspects of the invention in this regard has been made in a relatively general sense, except that positional control of an object has been assumed and aspects of a two-dimensional system have also been adopted. The invention could be applied in other embodiments, including some in which the controlled variable is not a position and does not directly relate to a position.

The Inventors of the present application contemplate that the control method, the system and the device described here be used in a variety of applications. Nevertheless were the control method, the system and the device to be described here, in fact for certain Advantages developed and have certain advantages when they be applied in a magnetic disk drive system. embodiments of the present invention be used advantageously to determine the position of a reading and write head with respect to a predetermined position or in Reference to a track or other sets of data storage locations to control. More preferably, these embodiments lead Control functions at least partially using an observable measure of the system to be controlled, a signal that comes from boundary conditions or an error information characteristic of on the hard disk drive stored data is derived from. Aspects of the present The invention will be with particular reference to these intended and particularly advantageous environments.

It should be noted that the environment of the intended magnetic disk drive is different from the conventional magnetic disk drive systems which control a head position based mainly on servo information such as those in the 9A & 9B are shown. The preferred embodiment of the magnetic disk drive provides a conventional signed and linear position error signal derived from the servo information stored in the sector servo parts stored in the sector servo 9A & 9B are shown. Control functions may be based on basics of position error signals derived from the observable information stored in servo parts such as those at A, B, C, D in the 9A & 9B are shown executed. Control functions are preferably performed based on observable information available at a rate substantially greater than the rate at which the servo part information is available.

The preferred operating environment conducts position error signals in one more continuous than is conventional in a disk drive, that uses a servo information only from servo burst patterns consists. Servo burst patterns of the type illustrated provide information not often as once per data block and more typically every three to five data blocks ready. Preferably, a suitable disk drive system which provides implements the aspects of the invention, observable information about position error signals frequently than once per data block ready. In particularly preferred embodiments could the observable position error signal from the data contained in a Block or from the signals representing that data, are derived and the position error signal is at least several Times during currently available, in which a data block is read from the disk. Most preferred will that extra Position error information is available from an observation, if the data is from the surface read the disk so that the position of the head during read operations can be controlled.

Especially preferred embodiments The present invention derives a position error signal the data stored on the disc, e.g. through evaluation of errors in the data or deviations from boundary conditions, which are contained in the data. Such boundary conditions, the Derivation of error signals on data based on boundary conditions and other aspects of a suitable and currently preferred operating environment be in the co-pending Application Serial No. 09/187770, by Alwin M. Despain et al., "Apparatus for developing a dynamic signal from data in a magnetic disc drive and method ", submitted on 06 November 1998 and jointly owned by the applicants, described. Despain application describes a control system which is able to control functions based on error or Execute constraint information, derived from data signals of the type that reside on the storage surface of a Plate could be stored.

The despain application describes position control functions that are performed, at least in part, using a position error signal that represents errors derived from data read from the disk. Two different position error signals, consistent with the teachings of the Despain application, each potentially capable of being used as an observable variable to control the position of a head of the disk drive, are disclosed in U.S. Patent Nos. 5,496,066 13 shown. It should be noted that the curves shown are highly idealized. For example, the illustrated curves are shown as being slightly asymmetrical with respect to the nominal track center. Empirical and theoretical data suggest that the curves are not generally symmetric and could be considerably less symmetrical than shown. The illustrated curves are also shown as monotone. Empirical data to date have not shown that the curves are strictly monotonic, as noise signals may be significant enough to encode the data in the 14 to modulate. The inventors of the present invention and the inventors listed in the Despain application have nevertheless been able to provide both modeled and actual disk drives based on the non-idealized error information corresponding to the type of data that is described in U.S. Patent Nos. 4,766,759 14 are shown to control.

It is important to note that the Despain application describes embodiments of a disk drive employing two read elements that collectively detect both a magnitude and a gradient of a position error signal. In effect, the two read elements detect two different position error signals corresponding to two slightly different displacements, which makes it possible to directly determine the sign of a head to the track displacement and a knowledge of the direction in which the head must be moved to correct for the displacement. allowed. Because the two read elements provide a gradient of the position error signal that can indicate the sign or direction for correction, the control method described herein is not necessary for the operation of such two read element headers, although aspects of the current control method which are useful. Aspects of the present invention have the greater advantage, for example, if gradient information is not available. Such a situation occurs when the head of the disk drive has a single read element and the observable variable has a functional relationship, such as that generally found in FIG 13 is shown.

The outermost curve in 13 is identified as having a bit error rate ("BER") as a function of the shift from a nominal track center or track to an optimum read position. The illustrated bit error rate is a position error signal which can be observed substantially continuously during data read operations and from the data recoverable from the disk sufficiently quickly to be used by a control system for controlling the relative position of the head with respect to a data track will be derived. It should be noted that there are a number of different possible measures of bit error rates and other functional relationships could be observed. The figure shows that the bit error rate changes as an absolute value function for positive and negative head shifts from the nominal track reading position. The figure also shows that the bit error rate is too small to be shown for small shifts and so could not currently be useful as an observable variable for a control system. For large displacements, however, the bit error rate is a strong function of the displacement and can be used for these large displacements to control the head position.

A study of the graphs of 13 suggests that the read head should be held at a nominal preferred reading position with respect to a track for reading data from the track with minimal errors. The asymmetry of the graphs, as well as the typical track margins, which are even allowed for a well described track (shown in 11 ) make it clear that the preferred track reading position need not be the middle of a track. It is nonetheless convenient to discuss the preferred reading position as being the center of the track that convention is often used in the technical field. A desired position for reading data from a track could be referred to herein as a nominal position, or sometimes as occurring along the centerline of the track. This use is intended to refer to a minimum fault location with respect to the lane, which of course results from the practice of certain aspects of the present invention. It was too 8th and the 9A & 9B It should be noted that the nominal head position for reading data differs from the nominal head position for writing data to a track.

The inner curve, which is called a "measure of error" in 13 is preferably preferred in a control system according to the present invention, because even smaller shifts dx lead to observable changes in the error measure. For example, the error measure illustrated could reflect an error that would result from multiple measures of signal quality, channel quality, or decoding accuracy generated within a Viterbi or other Trelles-type decoder that implements a Partial Response Maximum Likelihood (PRML) decoding function , is derived. A derivation of this particularly advantageous type of position error signal will be discussed in more detail below. 14 presents a more realistic set of error observations than the idealized graphs of 13 The vertical axis represents an average error over one thousand data points as a function of microinch track position. The origin of the track position axis is arbitrarily set on a page. It should be understood that observations from the illustrated data set could be used as an input to the control system, as long as the noise signals are properly taken into account by the control system in making decisions, although the illustrated graph is not straightforward.

The bit error rate, like the one in the outer BER curve of the 13 can be used as an observable position error signal for a control system to control the position of the head with respect to a data track. The other two error signals included in the 13 & 14 can be used in the same way. The illustrated position error signals have a generalized functional absolute value relationship to the displacement of the head from its nominal position. As such, detection of a given value of the bit error rate may indicate a positive or negative displacement of the head with respect to its nominal optimum track reading position. Preferred embodiments of the present invention use the absolute value control method to discriminate between initial positive and negative head shifts for those situations in which the sign of a shift corresponding to an observed position error signal is ambiguous.

There are a variety of head-to-track positioning operations that need to be controlled in a typical disk drive, including track seek operations and a track following operation and during both read and write operations. A track seek operation involves moving the head from a nominal read or write position with respect to one track to a nominal read or write position with respect to another track. Typically, a track seek operation is a relatively slow operation in which the head is moved over a number of tracks to reach the destination track. Because of the manner in which track seek operations are currently practiced, the large shift portions of the track seek operations are appropriately controlled and enabled by the servo information included in the 9A & 9B is shown. On the other hand, the absolute value control method is typically important in controlling the latter stages of track seek operations for practical disk drive designs. The absolute value control method is observed to provide settling times, as in 23 shown to improve significantly.

The Absolute value control method is used in tracking operations and in particular for tracking for read operations. A tracking for one Reading involves keeping the head above the Track being read, and preferably holding the head at or near a specific position relative to a nominal reading position. traditionally, this is done by adjusting the radial position of a head from the Sector advance information and then try the head at this radial To hold position while Data from the three to five Blocks that typically positioned between servo wedges. Preferred embodiments of present invention conduct error or other information off and use the data stored on the disk this information to control the position of the head during the read operation.

In accordance with particularly preferred implementations of the present invention generates a preferred control system position error signals (PES) from data on the surface a disk are stored at a significantly higher rate as the rate at the conventional Sector information available is. In certain present commercially available Systems is a sector information at a rate of about 15 kHz available. In a preferred implementation, the position error signal based on an error or constraint information characteristic The data generated at sequences as high as 300 kHz. Of course, future systems could are expected to provide characteristic servo patterns and data-derived servo patterns at different rates than discussed here. Especially This position error information is combined with high frequency with sector servo PES information generated from sector servo wedges at a much lower frequency than 15kHz, used to get a position error signal to create. To do this, the control system is preferably designed to to respond to low-frequency position error signals that have a sign or carry direction information, and position error signals a higher Frequency that have an absolute value characteristic that no Provide sign or direction information to address. Such a system could used to drive the hard disk drive at a frequency of position error signals between the lower frequency and the upper frequency, for example, 60 kHz, to control.

The Availability position error signals at considerably higher frequencies than conventionally derived position error signals enables a servo control system, which more accurate actuator inputs which produces a faster response, a smaller one Calming time and better rejection properties as those conventional ones Control systems that only lower the sector information with lower Use frequency derived from sector servo wedges that Information in a conventional low rate such as 15kHz.

The Absolute value aspects of the control method according to the present invention have to Do not be active at all times while viewing data from a disk recovered become. In other words, there are cases where the tax system with good certainty can know the sign of a shift, which generates an observed position error signal. For example there are many cases if the past history of head position and head movements an exact indication about the current one Sign or the direction of the shift provides. Exactly then when the sign of the displacement is needed both as is also ambiguous, are the absolute value aspects of the present Invention particularly useful. The higher Rate at which position error signals in accordance with aspects of this invention are available, is desirable independently whether the sign or the direction of the shift is known or not.

How the hard disk drive control system can have accurate information about the direction of displacement of a read head is in 15 schematically illustrating the path of a head, which is moved over the nominal target position within the target track in a position. The nominal target posi tion can correspond to the minimum of the "phase measure" position error signal graph, which in 13 is shown and assumed to be the optimal position within a track for reading data. In the figure, the head path is displayed as a thick black line, the target or nominal head position is indicated as the target position on the y-axis, and shifts from the nominal head position are displayed along the y-axis. For example, a shift of ε in the positive direction along the y-axis is displayed. Negative shifts are also shown for the head as a function of time. Because the only head movement that is relevant is along the radial direction, this is the only thing shown; Radial positions are indicated as a function of time, which extends along the x-axis.

In the presentation of 15 the head is displayed as being moved toward the target track position by a large positive offset. By the time the head comes within a distance of ε offset from the target position, the control system can safely assume that the head does not travel far enough into the next time interval to cross the target (eg, centerline) position the target track is moved translationally. Once the head is within a distance ε of the target (ie within ± ε of the target), the control logic begins to predict the position where the head should be at the next time interval. Alternatively, the control logic could predict the position of the head over other time intervals, such as constant. As discussed below, this prediction can be performed by a similar logic as used for the absolute value and other aspects of the control system. ε is a design parameter for the control system and hard disk drive system.

The control system continues to predict the head position at successive time intervals. The control logic monitors the head position predictions when the head is within the band of shifts that extend ± ε around the target to determine whether to expect. that the head crosses the target position. As long as the head does not cross the nominal track position of the target, the control system can continue using the historically correct positive sign for the displacement. If the head is predicted to pass the target position, the control system assumes that the sign of the absolute position error signal (such as the error metric of the 13 ) becomes ambiguous. The control system activates first and second estimators as soon as the sign of the shift becomes ambiguous to predict the future sign or the direction of displacement of the head.

Everyone the appraiser personified a model of the hard disk drive system that will determine the future head position based on the previous one State of the disk drive and input data accordingly to an assumption of the sign or the direction of the displacement of the head one next Time interval predicts. In the illustration, the estimator 2 contains the correct assumption of the sign or direction of the head one next Interval and so accurately predicts the actual path of the head. The appraiser 1 predicts an incorrect position that will eventually pass through the tax system is definitely identifiable as incorrect. After this the control system decides that the appraiser 1 was incorrect the condition of the appraiser 1 reset. The process for monitoring the position continues, with the head moving over the target position and the appraiser, which is correct, changes every time when the head is nominal Crossed finish position.

The dynamic processes the appraiser are based on accurate models of system dynamics and are such that, then, if the actual Signal available at the same sampling rate as the absolute signal, such as accurate estimates PES and system states would produce. Different appraisers become dependent from the available System model and used by its purpose. For example, linear Observers, non-linear observers, Kalman filters, input / output predictive models like ARMA (auto-regressive moving average; auto-regressive moving average) models, neural networks, systems with fuzzy logic, and the like can be used. Estimators can be implemented in software become independent whether in a general-purpose processor or in a more specialized one Digital signal processor, in hardware, or in combination of software and hardware. It is expected that different appraisers at are most suitable for are different systems to be controlled.

• • Wenn der Wert des Absolutwert-PES größer als ein bestimmter Pegel in Verbindung mit einer Verschiebung von ε entlang eine früheren Pfads ist, nimmt das Steuersystem an, dass sich das Vorzeichen der Verschiebung nicht geändert hat seit dem letzten Zeitintervall. Steuerfunktionen werden von dem Steuersystem in Übereinstimmung mit der Verschiebung, die aus dem beobachten Positionsfehlersignal abgeleitet wird, und dem Vorzeichen der Verschiebung bei dem früheren Zeitintervall ausgeführt.
• • Wenn der Absolutwert-PES kleiner als oder gleich wie ein bestimmter Pegel in Verbindung mit einer Verschiebung mit ε entlang des früheren Pfads ist und wenn die vorhergesagte Verschiebung des nächsten Schritts gleiche Vorzeichen wie die Verschiebung des vorangehenden Schritts aufweist, dann nimmt das Steuersystem an, dass das Vorzeichen der Verschiebung sich nicht geändert hat. Die vorhergesagte nächste Verschiebung kann unter Verwendung einer linearen oder nicht linearen Interpolation zwischen den letzten Signalabschätzungen erzeugt werden. Andere, mehr oder weniger komplizierte Verfahren, einschließlich von statistischen Verfahren, können verwendet werden, um die nächste Verschiebung vorherzusagen. Steuerfunktionen werden wiederum durch das Steuersystem in Übereinstimmung mit der Verschiebung, die aus beobachten Positionsfehlersignalen abgeleitet wird, und dem Vorzeichen der Verschiebung bei den früheren Zeitintervall ausgeführt.
• • Wenn der Wert des Absolutwert-PES kleiner also oder gleich wie ein bestimmter Pegel in Verbindung mit einer Verschiebung von ε entlang des früheren Pfads ist und wenn der vorhergesagte Wert der Verschiebung des nächsten Schritts ein entgegengesetztes Vorzeichen zu der Verschiebung des voranstehenden Schritts aufweist, dann wird das Vorzeichen einer nachfolgenden Verschiebung in der folgenden Weise abgeschätzt. Die Steuerlogik beobachtet ein Absolutwert-Positionsfehlersignal, welches auf erste und zweite mögliche Kopfverschiebungen abgebildet ist. Die Logik initiiert bzw. aktiviert einen ersten und zweiten Abschätzer, die als Anfangsbedingungen die ersten und zweiten möglichen Werte der Verschiebung nehmen. Verschiebungen und entsprechende Positionsfehlersignal werden vorhergesagt. Die Steuerlogik vergleicht die Positionsfehlersignale, die durch die Abschätzer vorhergesagt werden, mit dem gemessenen Positionsfehlersignal, um zu identifizieren, welcher Abschätzer das richtige Vorzeichen der Verschiebung einschließt. Sobald das richtige Vorzeichen abgeschätzt ist, wird das Vorzeichen mit der Verschiebung verwendet, um Steuerfunktionen auszuführen. Der Zustand des unrichtigen Abschätzers wird auf demjenigen des richtigen Abschätzers zurückgesetzt.

Control of the head position during tracking and other head positioning operations is accomplished by the control logic within a suitable control system. In summary, the control logic performs the following actions in translating a head to a track and, in some cases, causing the head to follow a track:

• If the value of the absolute value PES is greater than a certain level in conjunction with a displacement of ε along a previous path, the control system assumes that the sign of the Ver Shift has not changed since the last time interval. Control functions are executed by the control system in accordance with the shift derived from the observed position error signal and the sign of the shift at the earlier time interval.
• If the absolute value PES is less than or equal to a certain level in conjunction with a shift with ε along the previous path and if the predicted shift of the next step has the same sign as the shift of the previous step, then the control system assumes that the sign of the shift has not changed. The predicted next shift may be generated using linear or non-linear interpolation between the last signal estimates. Other more or less complicated procedures, including statistical methods, can be used to predict the next shift. Control functions are again performed by the control system in accordance with the shift derived from observing position error signals and the sign of the shift at the earlier time interval.
• If the value of the absolute value PES is less than or equal to a certain level in conjunction with a displacement of ε along the previous path and if the predicted value of the displacement of the next step has an opposite sign to the displacement of the previous step, then the sign of a subsequent shift is estimated in the following manner. The control logic observes an absolute position error signal, which is mapped to first and second possible head shifts. The logic initiates first and second estimators which take as initial conditions the first and second possible values of the displacement. Shifts and corresponding position error signal are predicted. The control logic compares the position error signals predicted by the estimators with the measured position error signal to identify which estimator includes the correct sign of the displacement. Once the correct sign is estimated, the sign is used with the displacement to perform control functions. The state of the incorrect estimator is reset to that of the correct estimator.

As soon as the sign of the shift has been estimated, is the estimate of the Signal substantially equal to either the output of the estimator, the the estimated one Sign or to the input of the estimator. always then, if an estimate the signal is carried out becomes the condition of the appraiser, the opposite of the estimated Sign corresponds, reset or, in other words, equal to the state of the estimator, the the estimated one Sign corresponds, set.

A detailed Description of aspects of one embodiment of the hard disk drive The present invention will now be presented. The specific embodiments are listed, to further implement specific aspects of the present invention Invention.

numerous changes from the basic embodiments are known and could in accordance with preferred embodiments of the present invention. For example, use the different embodiments Do not zones and set sector information with once at each three to five data blocks ready. As discussed in the background and as in the technical Area are known variations of such embodiments known, including from Disk drives, the zones and changing densities of sector information use at different radii.

16 is a simple block diagram of a hard disk drive that has a control system 110 according to the present invention. The servo control system 110 In the context of a hard disk drive (HDD) system having at least one rotatable magnetic disk 112 which stores data about one in accordance with conventional disk drive technology. The plate 112 has a plurality of substantially concentrically extending tracks 114 on, with only one of the tracks 114 in 16 for simplicity. The track 114 is through a read and write magnetic head 116 at the end of a head lift arm assembly 118 is appropriate, addressable. The arm 118 moves the head laterally relative to the surface of the plate 112 wherein it strongly moves the head along a radial path, and adjustable by an adjustment member or adjusting member 120 is positioned.

The plate 112 has a plurality of radially extending sector servo wedges 122 on, in an angular relationship at equal intervals over the storage surface of the plate 112 are arranged. Each of the sector servo wedges 122 provides a servo information, like the ones in the 9A & 9B shown and described above. When the plate is relative to the head 116 turns, then reads the head 116 the sector information out when the wedges 122 pass. The head and read channels of the disk drive circuitry derive a position error signal from the sector servo information that varies linearly with shifts from a nominal track. The resulting position error signal represents both a magnitude and a sign or direction of the adjustment to be made. to make a correction for the detected shift, ready. Between adjacent pairs of servo wedges 122 reads the head 116 Data taken along the track 114 are recorded. As a result, the signals passing through the head 116 be detected or read, both the servo information of servo wedges 122 as well as the data along the track 114 between the sector wedges 122 to be recorded. The general structure of the data blocks stored between the sector servo wedges is described above 9 & 10 illustrated and described above with reference to these figures.

The signals through the head 116 are detected, are connected to a phase locked loop 124 which detects and establishes the clock signal for sampling the signals from the disk at the appropriate clock rate. The output of the phase locked loop 124 is on a detector 126 applied, which may also include a detector. The detector 126 decodes the data signals coming from the track 114 be read, and put the decoded data as an output 128 ready. The detector 126 also processes the signals coming from the track 114 to read an error signal at a position error signal estimation circuit 130 provide. The error signal may have a characteristic like that in the 13 and 14 The output of the position error signal estimation circuit generates, and may be as described in the application of Despain, which is indicated above and by reference is part of the present application 130 is preferably a control signal such as an analog voltage suitable for adjusting the position of the read head. The output of the circuit 130 gets to the adjustment 120 applied to the head / loading arm 118 and the included head 116 reposition as needed, so that the head 116 exactly the track 114 tracked.

The magnetic disk 112 , as well as the arm 118 , the head 116 , the adjustment element 120 and the phase locked loop 124 can be of a conventional construction. Most preferably, the detector includes 126 Modifications in accordance with the present invention. to facilitate the use of this detector within preferred implementations of a control system. An acceptable detector is described in the Despain Application and will be described in more detail below. In accordance with the invention, the detector operates 126 and the position error signal estimating circuit 130 for estimating the positional error of the head 116 at a rate substantially greater than that conventional for hard disk drive control systems. It does so by determining the actual linear position error each time the servo burst occurs within one of the servo wedges 122 through the head 116 is detected. Between the servo wedges 122 and if the head 116 the data along the track 114 reads, values of the position error signal are derived and processed and are used to determine values of the displacement to be corrected by the control system.

Typically, those are along the track 114 recorded data in a particular format that imposes a constraint on the data stored by the system. In the present example, the data is encoded using PRML (partial response maximum likelihood), which is particularly relevant to today's disk drive industry. The detector 126 includes an algorithm for decoding the data. The data, as recorded on the disc, has a repetitive timing pattern caused by the phase locked loop 124 is used to set up a reference clock. This decode and error signal generation process is similar to that described in the despain application or like that described with reference to FIG 24 is described.

In the current example, the rotational speed of the disk 112 and the distance of the servo wedges 122 such that the servobursts of the wedges 122 be initiated at the head 116 pass by at a frequency of 15 kHz. Of course, other rates will be characteristic of other system designs. Every time the servo wedges 122 on the head 116 pass, a linear signal of the actual position error is generated. Both the magnitude and the sign of such a signal are known because of the nature of the servo bursts within the servo wedges 122 to be recorded.

Between the servo wedges 122 gropes the head 116 Tracking error data at a rate much higher than the rate of occurrence of the servo wedges 122 on the head 116 is. In the current example, the data rate could be 240 kHz and could even be greater, such as at a rate of 300 kHz if desired and as dictated by the system requirements. The timing is such that a position error signal from the data within the track 114 is generated at a rate higher than the rate of occurrence of the position error signals by detecting the servo bursts within the servo wedges by several factors 122 be generated. Thus, for each string of four position error signals, one by one sector 122 The other three could be generated by the data along the track 114 are recorded. As a result, the position error control system operates at a rate of 60 kHz in the current example. This is a quadruple of the sampling rate provided by the sector servo wedges 122 used alone.

In accordance with the invention, a position error estimate is made using the absolute value type of position error that results from along the track 114 Recorded data signals can be derived, a much higher sampling rate possible. Aspects of the operation of the position error signal estimation circuit 130 will be described in more detail below in connection with the block diagram of 17 described. 17 schematically illustrates aspects of the logical operations performed by the servo control system 110 of the 16 and in particular the position error signal estimation circuit 130 be executed. 17 also illustrates aspects of the models underlying the control logic and the disk drive. The discussion of 17 is more general than presentation in many aspects 16 reflecting the fact that aspects of the control process are useful for systems other than hard drives. In the presentation of the 17 represents the servo mechanism 140 and the non-linear map 142 the hard disk drive other than the position error signal estimation circuit 130 The non-linear map 142 connects the observed error signal to the displacement of the head from its nominal position. The output from the logic circuit 154 of the 17 could be on the adjustment 120 ( 16 ) or could be more generally converted before being attached to the adjustment element 120 provided.

The control logic arrangement of 17 receives as inputs from the hard disk drive the measurements of its sampled output signals, including both the conventional linear servo information derived from the sector servo wedges and the absolute value servo information derived from an error or constraint information characteristic of the data stored on the hard disk could. Regardless of the source of absolute value servo information, the illustrated example provides linear servo information at a relatively low rate and absolute servo information at a much higher rate. Thus, the linear servo information is provided at a frequency of Xs kHz and the absolute value servo information is provided at a much higher frequency Xf kHz. The control system operates at a frequency of Xc kHz which is higher than Xs, but generally lower than Xf kHz. In the present example, the frequencies may be Xs = 15 kHz, Xf = 240 kHz and Xc = 60 kHz, but these values are only examples.

For the hard disk drive (Hard Disk Drive, HDD) of the example, y is the displacement or Deviation and f (y) is the observable PES corresponding to the Shift and generated from reading the data track. f (y) becomes as a continuous signal and in practice can f (y) be available at sampling frequencies such as 200 MHz. The servo control system is for that designed with a significantly higher frequency than the linear one Servo information frequency to work, because the operation with the higher Frequency better tracking and sedation features of preferred embodiments of the present invention.

It is not necessary to operate at the higher frequency of the absolute value servo information to obtain improved performance. In certain circumstances, it is not preferred to operate at the highest frequency because the circuitry needed is more complex and the resulting system may not be as stable as desired. It should be noted that an optimal control system for system scan information at a rate of 15 kHz is relatively different from optimal control system scan information at 60 kHz. Further, a control system capable of making corrections using information sampled at a rate of 60 kHz is not expected to be optimal for a system operating at a rate of 240 kHz Models of system dynamics in different cases in the control logic, which in 17 is shown used. The following with reference to 20 The control system described is partially chosen to operate at a frequency of 60 kHz, with certain of the servo information samples (one of 4) representing linear servo information having an observed magnitude and sign, and others of the servo information samples, (3 of 4). represent absolute value servo information available at a higher frequency but with an estimated sign. The absolute value servo information samples themselves are the result of several observations of the position error signal.

The arrangement of 17 includes a servo mechanism 140 which produces a head-to-track displacement y at intervals during normal operation. In the servo control system model of 17 represents the servo mechanism 140 The dynamic properties of the servomechanism and the response thereof in a mostly conventional manner. The control logic illustrated uses a time-invariant model for the dynamic properties of the servomechanism described with the following equations: X = AX + Bu, X (0) = X O y = CX (1) where X is the state vector, y is the scalar output, and u is the scalar input. The matrices A, B, C each have suitable dimensions with constant coefficients. This model is an approximation of a more complex model of dynamic servo properties, which in some cases is preferred and discussed below. A simple model that captures the dominant characteristics of the dynamic properties is preferred in other cases because a simple model results in a simpler and less computationally complex algorithm and a less expensive control design.

The output y of the servo mechanism 140 is an entrance to a non-linear maps 142 from y. Here, y is the displacement of the controlled variable which is not directly observable, and the non-linear map produces the position error signal which is observable. An example of the map 142 is in 18 and, generally, the error information graph included in the 13 & 14 are shown. The output y is converted to a non-linear function f (y) representing the error rate. A given value of f (y) could be generated by two possible values of y, which are y ⁺ and y ^- . The map 142 represents what happens in the disk drive to produce the position error signal f (y). As such, this card does not reflect logic circuitry but instead illustrates how the disk drive generates the observable variable that can be used as an indication of the displacement.

The block 144 in the arrangement of 17 is a sample and hold (S / H) device which samples and holds the value of f (y) at a sampling frequency of Xf kHz. The S / H device is typically implemented as a zero order hold (ZOH) and can be implemented in hardware or software. In the current example, Xf = 240 kHz, which means that samples of f (y) are available every T ₂₄₀ = 10 ^-3 / 240 seconds. The output of the S / H device 144 is f (y) j = f (jT 240 ) which is a sequence of steps with values f (y) _j at times t = jT ₂₄₀ , j = 1, 2, ..., in other words f (y) _j = f (y (t)) for jT ₂₄₀ ≤ t <(j + 1) T ₂₄₀ .

= y(kT₁₅) zu der Zeit t = kT₁₅, k = 1, 2, ..., d.h.

= y(t) für kT₁₅ ≤ t < (k + 1)T₁₅.A block 146 opposite to the block 144 is a sample and hold (S / H) device which samples and holds the value of y at a sampling frequency of Xs kHz. In the present example, Xs = 15 kHz. This means that a low frequency linear servo information component of y is available every T ₁₅ = 10 ^-3 / 15 seconds. The output of the S / H device is a sequence of steps with values

= y ( k T ₁₅ ) at the time t = k T ₁₅ , k = 1, 2, ..., ie

= y (t) for k T ₁₅ ≤ t <( k + 1) T ₁₅ .

The sample and hold device 144 is with an inverse card 148 coupled. The inverse card 148 represents the inversion of the function by the non-linear map 142 is provided and the inversion of the function, which in 18 plotted, ready. Whereas the non-linear map 142 provides the error rate f (y) for the scalar value y, represents the inverse map 148 the scalar shift values y _j ⁺ and y _j ^- in response to the error rate f (y) _j . In particular, the inverse card uses 148 the sampled values of f (y) _j and the non-linear mapping to determine the corresponding sets of possible values of y at time intervals t = jT ₂₄₀ . As in 18 these values are shown by y _j ^-, y _j ^+, respectively. For a given value of f (y) _j, there are two possible values of y at t = jT ₂₄₀ , ie y j - = y - (jT 240 ) y j + = y + (T 240 ).

• • x ^_1,j = x ^₁(jT₂₄₀) der Wert des Zustands x ^₁ zu der Zeit t = jT₂₄₀
• •
  

An exit of the inverse map 148 and the output of the sample and hold device 146 become to a first estimate 150 (Estimator 1) coupled. As a practical course of action are the inverse map 148 and the estimators implemented as software, although aspects could be implemented in hardware. The estimator 1 is described by the following equations, in which the letters A, B, C indicate discrete forms of the matrices discussed above: x ^ 1, j + 1 = A 240 x ^ 1, j + B 240 u k + K 240 (C 240 x ^ 1, j - y + j ) x ^ (jT 240 ) = IC 1 y ^ 1, j = C 240 x ^ 1, j With

• • x ^ _{1, j} = x ^ ₁ (jT ₂₄₀ ) is the value of the state x ^ ₁ at the time t = jT ₂₄₀
• •
  

nicht verfügbar ist, wird die positive Verschiebung, die aus beobachteten PES erzeugt wird, unter Verwendung der inversen Karte 148 y_j ⁺ verwendet. Wegen der Konvention der Notation ist y_j ⁺ der Wert von y_j = y(jT₂₄₀), wenn y_j positiv ist oder y_j ^–, wenn y_j negativ ist. Da aus der Messung f(y)_j zwei mögliche Werte y_j ⁺, y_j ^– erhalten werden, schätzt der Abschätzer 1 die Werte von y_j ⁺ ab und ein zweiter Abschätzer nimmt den Wert von y_j ^– an. Der zweite Abschätzer (Abschätzer 2) ist mit einem Block 152 in 17 dargestellt.

• • A₂₄₀, B₂₄₀, C₂₄₀ sind die entsprechenden Matrizen, die durch Diskretisierung des Servomechanismus 140 mit einer Abtastperiode T₂₄₀ = 10^–3/240 Sekunden erhalten wird. Sie werden mit den folgenden Beziehungen gegeben:
• •
  
• • u_k = u(kT₆₀) ist der Stellglied-Eingang zu dem Servomechanismus 140 bei t = kT₆₀, k = 1, 2, ... durch den Controller zu erzeugen, wobei T₆₀ = 10^–3/60 Sekunden die von der Regelschleife verwendete Abtastperiode ist.
• • K₂₄₀ ist eine Designmatrix, die so gewählt ist, dass die Matrix A₂₄₀ + K₂₄₀C Eigenwerte mit einer Größe kleiner als eins und nahe zu Null aufweist Der Ort der Eigenwerte beeinflusst die Geschwindigkeit des Ansprechverhaltens des Abschätzers dahingehend, dass das System schneller ist, wenn sie nahe zu Null sind.
• • x ^₁(jT₂₄₀) = IC₁ zeigt die Initialisierung der Sequenz zur Zeit t = jT₂₄₀, für irgendeine ganze Zahl j bis IC₁, erzeugt durch den nicht linearen Logikblock 154, der mit den Ausgängen der Abschätzer 150 und 152 gekoppelt ist.
• • y ^_1,j = y ^₁(jT₂₄₀) ist der Ausgang des Abschätzers 1.

Ideally should y _j ⁺ equal to y _j = y (jT ₂₄₀ ). In this case it can be established that x ^ _{1, j} → _xj , y ^ _{1, j} → _yj ; since j → ∞, where x _j = x (jT ₂₄₀ ) In other words, the state of the estimator converges to the actual state of the system. Since y _{j is} not available, that is, it can not be observed, the estimated value y _k = y ( k T ₁₅ ). This is the actual measurement of y at t = k T ₁₅ (those intervals at which the linear servo information is available) and for the times at which

is not available, the positive shift generated from observed PES is obtained using the inverse map 148 y _j ⁺ used. Because of the convention of notation, y _j ^{+ is} the value of y _j = y (jT ₂₄₀ ) if y _{j is} positive or y _j ^- if y _{j is} negative. Since two possible values y _j ⁺ , y _j ^- are obtained from the measurement f (y) _j , the estimator 1 estimates the values of y _j ⁺ and a second estimator takes the value of y _j ^- . The second estimator (estimator 2) is one block 152 in 17 shown.

• • A ₂₄₀ , B ₂₄₀ , C ₂₄₀ are the corresponding matrices created by discretization of the servo mechanism 140 with a sampling period T ₂₄₀ = 10 ^-3 / 240 seconds. They are given with the following relationships:
• •
  
• • u _k = u (kT ₆₀ ) is the actuator input to the servo mechanism 140 at t = kT ₆₀ , k = 1, 2, ... by the controller, where T ₆₀ = 10 ^-3 / 60 seconds is the sampling period used by the control loop.
• K ₂₄₀ is a design matrix chosen such that the matrix A ₂₄₀ + K ₂₄₀ C has eigenvalues smaller than one size and close to zero. The location of the eigenvalues affects the speed of the estimator's response to make the system faster is when they are close to zero.
• • x ^ ₁ ( j T ₂₄₀ ) = IC ₁ shows the initialization of the sequence at time t = j T ₂₄₀ , for any integer j to IC ₁ , generated by the non-linear logic block 154 , with the exits of the appraisers 150 and 152 is coupled.
• • y ^ _{1, j} = y ^ ₁ (jT ₂₄₀ ) is the output of the estimator 1.

It should be noted that if, for each j, y _{j is} positive, meaning that y _j ^{+ is} the value of y that produced f (y) _j , then the estimator 1 assumes that y ^ _j → y _j is when j → ∞ goes.

• • A₂₄₀, B₂₄₀, K₂₄₀, C₂₄₀, u_k sind wie beim Abschätzer 1 definiert.
• • x ^₂(jT₂₄₀) = IC₂ zeigt die Initialisierung der Sequenz x ^_2,j zur Zeit t = jT₂₄₀ auf den Wert IC₂, der durch die nicht lineare Logik 154 erzeugt wird.
• • Der Ausgang des Abschätzers 2 ist y ^_2,j = y ^₂(jT₂₄₀) zur Zeit t = jT₂₄₀.

The appraiser 2, the block 152 in 17 is described with the following equations: x ^ 2, j + 1 = A 240 x ^ 2, j + B 240 u k + K 240 (C 240 x ^ 2, j - y - j ) x ^ 2 ( j T 240 ) = IC 2 y ^ 2, j = C 240 x ^ 2, j With

• • A ₂₄₀ , B ₂₄₀ , K ₂₄₀ , C ₂₄₀ , u _k are defined as in estimator 1.
• • x ^ ₂ ( j T ₂₄₀ ) = IC ₂ shows the initialization of the sequence x ^ _{2, j} at time t = j T ₂₄₀ to the value IC ₂ , by the non-linear logic 154 is produced.
• The output of the estimator 2 is y ^ _{2, j} = y ^ ₂ (jT ₂₄₀ ) at time t = jT ₂₄₀ .

= y(kT₁₅) verwendet wird.The estimator 2 assumes that the value of y in the inverse map of f (y) is negative and uses y _j ^- for the possible correction value of y at t = jT ₂₄₀ except for cases of time where j = 16 k ie the linear servo information is available, in which case the measured value of y,

= y ( k T ₁₅ ) is used.

= y(kT₁₅) und y_j ⁺, y_j ^–, wobei j, k die Zeitindices sind, die den Abtastfrequenzen 240 kHz (Servo mit hoher Frequenz) bzw. 15 kHz (lineare Servo) entsprechen. Der Ausgang des Logikblocks 154 ist die Abschätzung des Signals y bei t = kT₆₀ und wird mit y ^_k = y ^(kT₆₀) bezeichnet, wobei k der Zeitindex entsprechend zu der Abtastsequenz von 60 kHz ist. Die Beziehung der Indices ist in 19 gezeigt.The inputs to the non-linear logic block 154 are y ^ _{1, j} = y ^ ₁ (jT ₂₄₀ ), y ^ _{2, j} = y ^ ₂ (jT ₂₄₀ ),

= y ( k T ₁₅ ) and y _j ⁺ , y _j ^- , where j, k are the time indices corresponding to the sampling frequencies 240 kHz (high frequency servo) and 15 kHz (linear servo), respectively. The output of the logic block 154 is the estimate of the signal y at t = kT ₆₀ and is denoted by y ^ _k = y ^ (kT ₆₀ ), where k is the time index corresponding to the sampling sequence of 60 kHz. The relationship of the indices is in 19 shown.

The index k corresponds to the time t = k T ₁₅

The index k corresponds to the time t = kT ₆₀

The index j corresponds to the time t = jT ₂₄₀

• 1. Wenn j = 16k ist, dann ist ŝ_k = sgn(
  
  ), mit anderen Worten, wenn j = 16k ist, bedeutet dies, dass der Zeitpunkt t = kT₆₀ ist Da der Wert der linearen Servoinformation
  
  = y(kT₆₀) zu diesem Zeitpunkt verfügbar ist, ist das Vorzeichen von
  
  bekannt und wird zur Verwendung durch das Steuersystem ausgegeben.
• 2. Wenn y_j ⁺ oder f(y_j) bei j = 4k größer als ε% der Spurbreite ist, d.h., der Kopf wird ausreichend weit weg von der nominellen Zielspurposition verschoben, dann gilt ŝ_k = ŝ_k-1. Der Anfangswert der Vorzeichenabschätzung wird so gewählt, dass er das tatsächliche Signal bei dem letzten Abtastintervall ist. Mit anderen Worten, wenn der Wert von y_j ⁺ oder f(y_j) > ε% ist, wobei ε > 0 eine Zahl ist, die den Rauschpegel anzeigt, wird keine Vorzeichenänderung von dem vorangehenden Schritt angenommen und deshalb bleib das Vorzeichen das gleiche. Der Wert von ε wird z.B. so gewählt, dass er geringfügig höher als der Rauschpegel an dem Ausgang y ist.
• 3. Wenn y_j ⁺ oder f(y_j) bei j = 4k kleiner als ε% ist, dann gibt es zwei Fälle:
• 3a. Fall I Der Wert von y _pk von y_k wird unter Verwendung des Filters y ^pk = 2y ^k – y ^k-1 vorhergesagt, das unter Verwendung einer linearen Interpolation zwischen y ^_k-1 und y ^_k-2 entwickelt wird. Wenn sgn(y ^pk) = sgn(y ^k-1)ist, dann gilt: ŝ_k = ŝ_k-1. Mit anderen Worten, wenn das Vorzeichen des interpolierten nächsten Schrittwerts von y_k so vorhergesagt wird, dass es das Vorzeichen der Abschätzung y ^_k-1 in dem vorangehenden Schritt ist, dann wird keine Vorzeichenänderung erwartet und deshalb ist ŝ_k = ŝ_k-1.
• 3b. Fall II Wenn sgn(y ^_pk) ≠ sgn(y ^_k-1) ist, dann gilt:
  
  und IC1 = x ^2,j wenn ŝk = –1 und j = 4k IC2 = x ^1,j wenn ŝk = +1 und j = 4k

ŝ _k is the estimate of the sign s _k of y _k = y (kT ₆₀ ). The functions of the non-linear logic circuit 154 are described as follows:

• 1. If j = 16 k is, then ŝ _k = sgn (
  
  ), in other words, if j = 16 k is, this means that the time t = k T ₆₀ is the value of the linear servo information
  
  = y ( k T ₆₀ ) is available at this time, the sign of
  
  and is issued for use by the control system.
• 2. If y _j ⁺ or f (y _j ) at j = 4k is greater than ε% of the track width, ie, the head is displaced sufficiently far away from the nominal target track position, then ŝ _k = ŝ _k-1 . The initial value of the sign estimate is chosen to be the actual signal at the last sampling interval. In other words, if the value of y _j ⁺ or f (y _j )> ε%, where ε> 0 is a number indicating the noise level, no sign change is taken from the previous step, and therefore the sign remains the same , For example, the value of ε is chosen to be slightly higher than the noise level at the output y.
• 3. If y _j ⁺ or f (y _j ) is less than ε% at j = 4k, then there are two cases:
• 3a. Case I The value of y _pk of y _k is made using the filter y ^ pk = 2y ^ k - y ^ k-1 predicted, which is developed using a linear interpolation between y ^ _k-1 and y ^ _k-2 . If sgn (y ^ pk ) = sgn (y ^ k-1 ) is, then: ŝ _k = ŝ _k-1 . In other words, if the sign of the interpolated next step value of y _k is predicted to be the sign of the estimate y _k-1 in the previous step, then no sign change is expected and therefore ŝ _k = ŝ _k-1 ,
• 3b. Case II If sgn (y ^ _pk ) ≠ sgn (y ^ _k-1 ), then:
  
  and IC 1 = x ^ 2, j if ŝ k = -1 and j = 4k IC 2 = x ^ 1, j if ŝ k = +1 and j = 4k

In other words, if the sign of the predicted value of y _{k is} different from the sign of the estimate y _k-1 in the previous step, then the sign of y _k is expected to change in those time intervals. In this case, the comparison discussed above is performed to determine how close the outputs of the estimators are to the assumed correct values of y _j ⁺ , y _j ^- . For example, if y _{j was} positive, then y _j ^{+ was} the correct value, and if y _{j was} negative, then y _j ^{- was} the correct value. Therefore, the estimator that used the correct value will have an estimated output that is closer to the correct value. If eg (y ^ 2 1, j - y + j 2 ) 2 <(y ^ 2 2, j - y - j 2 ) 2 is satisfied at j = 4k, then this means that the estimator 1 is the one that has used the correct value, which in turn implies that y _{j is} expected to be positive, in other words, ŝ _k = 1. If the above inequality does not is satisfied, then this would imply that the estimator 2 is the correct one and ŝ _k = -1. The incorrect estimator must have its state reinitialized to that of the correct estimator to prevent the previous estimation errors from propagating. Therefore, the initial conditions IC ₁ , IC ₂ for the estimators are selected as indicated above.

1. Estimate y ^ _k

The estimate of y ^ _k is formed as follows:

, der tatsächliche abgetastete Wert von y verfügbar ist und verwendet wird, weil die lineare Servoinformation sowohl die Größe als auch das Vorzeichen für die Verschiebung bereitstellt. Wenn nicht, dann wird der Ausgang des Abschätzers zu allen anderen Zeitintervallen verwendet, in Abhängigkeit von dem Wert von ŝ_k. Wenn ŝ_k = 1 ist, dann wird der Ausgang des Abschätzers 1 verwendet, und wenn ŝ_k = –1 ist, dann wird der Ausgang des Abschätzers 2 verwendet.In other words, the estimate y ^ _k = y ^ (kT ₆₀ ) is the operation at the time (k = 4 k ), in which

For example, the actual sampled value of y is available and used because the linear servo information provides both the magnitude and sign for the displacement. If not, then the output of the estimator is used at all other time intervals, depending on the value of ŝ _k . If ŝ _k = 1, then the output of the estimator 1 is used, and if ŝ _k = -1, then the output of the estimator 2 is used.

The estimated value of y ^ _k of y at t = kT ₆₀ is used to construct a controller to operate at a frequency of 60 kHz. The construction of a suitable servo system is conventional and may be constructed, for example, in accordance with the general teachings of Chapter 14 of Franklin et al., Digital Control of Dynamic Systems, 649-687 (1998), this chapter describing the so-called Workman model. The block diagram of the controller together with the blocks for generating y _k are in 20 shown. The illustrated control system, including the estimators, the non-linear logic, and the controller, is preferably implemented using software, eg, within a digital signal processor, as presently conventional in the art.

• • A₆₀, B₆₀, C₆₀ die diskreten Zeitmatrizen des dynamischen Modells (1) des Servomechanismus abgetastet bei 60 kHz sind, mit anderen Worten:
  
• • G₆₀ eine Matrix ist, die so gewählt ist, dass die Eigenwerte von A₆₀ + G₆₀C₆₀ eine Größe kleiner als Eins und nahezu Null haben.

The arrangement of 20 includes a controller observer 160 representing the observer using the estimated value of y to generate an estimate of the state x of the model of the servo mechanism at times t = kT ₆₀ , where T ₆₀ = 10 ^-3 / 60 seconds. The controller 160 is described with the following equations x ^ k + 1 = A 60 x ^ k + B 60 u k + G 60 [C 60 x ^ k - y ^ k ] With:

• • A ₆₀ , B ₆₀ , C _{60 are} the discrete time matrices of the dynamic model (1) of the servo mechanism sampled at 60 kHz, in other words:
  
• • G _{60 is} a matrix chosen so that the eigenvalues of A ₆₀ + G ₆₀ C _{60 have} a size smaller than one and close to zero.

• • K₆₀ eine Controller-Verstärkungsmatrix ist, die so konstruiert ist, dass die Größen der Eigenwerte der Matrix A₆₀ – B₆₀K₆₀ kleiner als Eins sind und in anderen Bereichen, um die Steuerziele zu erfüllen. Der Wert von K₆₀ kann unter Verwendung eines H_∞ Frequenzdomänen-Ansatz oder anderen geeigneten Verfahren gewählt werden. Für den Fall eines Festplattenlaufwerks (HDD) ist y die Verschiebung und die Controllerverstärkung K₆₀ wird vorzugsweise dafür ausgelegt, um die folgenden Ziele zu erfüllen:
• • Erreichen einer Spursuche in weniger als 1,5 Millisekunden. Eine Spursuche wird als vollständig angenommen, wenn die Verschiebung weniger als 3% der Spurbreite ist.
• • Die Systembandbreite der geschlossenen Schleife sollte nicht 1,5 kHz überschreiben, um eine Anregung der Hochfrequenzmoden des Systems zu vermeiden.
• • Die maximale Spitze der komplementären Empfindlichkeit und der Empfindlichkeitsfunktionen sollte 3–4 dB nicht überschreiten, um eine Verstärkung von Störungen zu vermeiden, und ausreichende Systemberuhigungszeit und Systemrobustheit zu ermöglichen. Die Funktionen der komplementären Empfindlichkeit Hcs und der Empfindlichkeit Hs werden jeweils wie folgt definiert:
  
  wobei K(s) die Controller-Übertragungsfunktion bezeichnet und H(s) die Übertragungsfunktion der dynamischen Eigenschaften des Festplattenlaufwerks bezeichnet. Es sei darauf hingewiesen, dass das bestimmte Servosteuerungsmodell für jedes Plattenlaufwerk angepasst werden wird und im allgemeinen für jedes bestimmte Laufwerk, um ein optimales Betriebsverhalten zu erhalten.

The arrangement of 20 includes the block 162 which outputs a control signal u _k in response to the input of an estimate of the variable x. The output control signal is given as follows: u k = -K 60 x ^ k With:

• K _{60 is} a controller gain matrix designed so that the eigenvalues of matrix A ₆₀ -B ₆₀ K _{60 are} less than unity and in other ranges to meet the control objectives. The value of K ₆₀ may be chosen using a H _∞ frequency domain approach or other suitable method. In the case of a hard disk drive (HDD), y is the displacement and the controller gain K ₆₀ is preferably designed to meet the following objectives:
• • Achieving a track search in less than 1.5 milliseconds. A track search is considered complete when the displacement is less than 3% of the track width.
• • The closed loop system bandwidth should not overwrite 1.5 kHz to avoid exciting the system's high frequency modes.
• • The maximum peak of the complementary sensitivity and sensitivity functions should not exceed 3-4 dB to avoid amplification of noise and allow for sufficient system settling time and system robustness. The functions of the complementary sensitivity Hcs and the sensitivity Hs are respectively defined as follows:
  
  where K (s) denotes the controller transfer function and H (s) denotes the transfer function of the dynamic characteristics of the hard disk drive. It should be appreciated that the particular servo control model will be adapted for each disk drive, and generally for each particular drive, for optimum performance.

wobei Y(s), U(s) die Laplace Transformationen der Kopfposition bzw. des Stellgliedeingangs bezeichnen. Die Parameter des Modells sind in der Tabelle 1 aufgelistet.In the simulations, in the graphs with the 21 and 22 The following transfer function model describing the dynamic properties of the hard disk drive was used:

where Y (s), U (s) denote the Laplace transformations of the head position and the actuator input, respectively. The parameters of the model are listed in Table 1.

TABLE 1: SYSTEM PARAMETERS

und deshalb können die dynamischen Eigenschaften des Festplattenlaufwerks durch das folgende Übertragungsfunktionsmodell approximiert werden:

was die Dynamik der reduzierten Ordnung darstellt, die für Steuerzwecke verwenden werden soll und auch mit Gleichung (2) des Servomechanismus 140 beschrieben wird. Der Servocontroller ist auf Grundlage des Modells der reduzierten Ordnung konstruiert, wird aber unter Verwendung des Systems der vollen Ordnung symbolisiert und analysiert.A careful consideration of the transfer function model given by equation (2) above for the servomechanism 140 of the 17 is suggested that the mode corresponding to the first resonant frequency of 70 Hz is preferably neglected for control design purposes. The next resonant frequencies are at 2.2 kHz, 4 kHz and 9 kHz. If the closed-loop bandwidth is kept smaller than 1.5 kHz, then the modes corresponding to these frequencies can be neglected for control purposes. These considerations lead to the following approximation:

and therefore the dynamic properties of the hard disk drive can be approximated by the following transfer function model:

which represents the dynamics of the reduced order that should be used for tax purposes and also with Equation (2) of the servomechanism 140 is described. The servo controller is constructed on the basis of the reduced order model, but is symbolized and analyzed using the system of full order.

21 shows plots of the actual displacement and their estimation as determined by the in 17 shown arrangement generated. A plot 170 is from the actual displacement, while a plot 172 of whose estimation is. As shown, the plots are essentially identical and can not be distinguished in this view, even if a pulse 174 is applied to the systems. This indicates a high degree of accuracy in the predicted positions, as well as the dynamic properties, for the described absolute value control ethology.

22 FIG. 12 shows the head shift as a function of time during and after a fault that shifts the head using the controller having the same structure as that described above for two different cases. In a case representing conventional systems, no estimation of the displacement takes place and the linear servo signal at 15 kHz is used. The controller also works at the 15 kHz frequency. The upper plot 176 in 22 shows the head shift as a function of time for the 15 kHz case. The lower plot 178 shows the head shift as a function of time when the estimated value of the PES is sampled at 240kHz and the controller operates at 60kHz, as in FIG 20 shown. As illustrated, the disk drive, controlled in accordance with aspects of the present invention, exhibits significantly better responsiveness and stability compared to conventional systems.

23 FIG. 12 shows the head shift as a function of time after a track search using the above-described controller for the same two cases described with reference to FIG 22 were discussed. In a case representing conventional systems, no estimation of the displacement takes place, and the linear servo signal at 15 kHz is used for position control. The controller works at the 15 kHz frequency. The y-axis displacement is measured in units of track width and the x-axis time is measured in milliseconds. The upper plot 180 in 23 shows the head shift as a function of time for the 15 kHz case. The lower plot 182 shows the head shift as a function of time when the estimated value of PES is sampled at 240kHz and the controller operates at 60kHz, as in FIG 20 shown. As illustrated, the disk drive controlled in accordance with aspects of the present invention provides significantly better responsiveness and stability compared to conventional systems. For example, the faster response and the faster settling time are shown in the generation of a search time improvement on the order of half a millisecond.

As discussed above, is the preferred source for position error signals (PES) for a position control of a head with respect to a track in one Disk drive the data within the track itself. In particular For example, the preferred PES signals will be information about error characteristics the data or the data signals stored on the disk are derived. Another related preferred source of PES becomes constraint information characteristics the data or the data signals in connection with the data or Derived data signals themselves. The derivation of this type of error and restriction information is described in the aforementioned and incorporated Despain application.

24 illustrates what is commonly known as the read channel of a hard disk drive, which is typically embodied on a single chip, sometimes as an ASIC (Application Specific Integrated Circuits), sometimes as a specialized chip, and sometimes as a set of functionality within one Chips for digital signal processing (DSP chip). Individual circuit elements of the read channel chip are familiar to those of ordinary skill in the art and are described in patents such as US Patent No. 5,585,975, US Patent No. 5,754,353, and US Patent No. 5,961,658. Each of these patents relates to the structure and function of the circuit elements and accumulations of elements incorporated in 24 are shown. In most conventional disk drives, a read channel chip, which in many ways may be similar to the one in FIG 24 is used to receive and process the linear servo information provided in the sector servo wedges of conventional disk drives. For example, aspects of such use are described in US Pat. No. 5,825,579.

The in 24 The read channel shown is specifically designed to implement aspects of the present invention and the control methods and systems described in the Despain Application the. Several measures of quality or equivalent errors are made in the read channel chip 24 generated. Most preferably, at least one of these will be misjudged as an output from that in 24 provided chip provided. For example, one or more error measures could be provided to a corresponding one or more of the output pins or the output pins of the chip. Such an output port or pin is a particularly preferred source of the absolute value PES signals used in hard disk drive environments in particularly preferred implementations of the present invention.

Referring to 24 becomes the reading head 200 is identified with a shade to indicate that it is not on the read channel chip, but instead provides the signals that the read channel chip is processing. Signals coming from the read head 200 are output to a preamplifier 202 provided that performs a conventional broadband amplification. The output from the preamplifier 202 will be on the amplifier 204 of variable gain, the gain of which is adjusted so that the signals presented to the remainder of the circuitry have an amplitude within a range of acceptable amplitudes. The logic 206 Automatic Gain Control (AGC) detects signal levels and certain other signal characteristics from digitizing and equalization circuitry within the read channel chip, and from these signals directs the gain to the amplifier 204 with variable gain. The AGC logic 206 sets the gain on the digital-to-analog converter 208 ready, in turn, the gain to the amplifier 204 with variable gain supplies.

The signal output from the variable gain amplifier is analog and has peak voltages that are more accurate in the desired range. The variable gain amplifier has an analog equalizer 210 coupled. The analog equalizer 210 adjusts the signal level over a range of received frequencies to remove some of the frequency dependent effects of the read head and other portions of the read channel circuitry. After equalization, the signal is applied to the analog-to-digital converter 212 provided in accordance with a clock signal generated by the clock recovery circuitry 214 is provided to filter the specified analog signals. As in 24 includes the clock recovery circuitry 214 a Phase Locked Loop (PLL) detector for identifying and extracting timing information and a voltage controlled oscillator (VCO) for generating a clock signal suitable for the rest of the circuitry in the read channel chip.

The output of the analog-to-digital converter 212 gets to a second equalizer 216 which is used to further equalize the signal. In the illustrated circuit, the second equalization circuit is in the form of a finite impulse response (FIR) filter. The equalizer or equalization provided by the FIR filter 216 is adjusted in accordance with the detected signal levels to receive the data received by the circuit. This adaptive adjustment process is accomplished by adjusting the weight of the FIR filter using information provided by a slicer (separation unit). 218 is provided. The slicer 218 generates estimated samples by comparing individual samples with expected samples. For example, the slicer could compare each sample that is received to values set for logic signals of +1, 0, and -1, and assign to each sample the logical signal that most closely matches its value. These estimates are used to adjust the gain, set the clock recovery circuitry, and set the equalization process.

As part of the Slicer functions 218 The slicer generates a measure of channel quality by comparing each of the values received by the slicer received by the slicer with the value assigned by the slicer to that sample. For example, a difference measurement may be calculated for each sample. A first sample provided at the slicer has a voltage VS. The slicer determines that the voltage VS is closest in value to the voltage V1 expected for samples corresponding to the logic value +1. The sample is preliminarily assigned the logic value +1, and the slicer can then generate an error measure for that sample equal to VS-V1. Alternatively, the error measure for this sample could be the absolute value of that difference or the square of that difference. Various statistical analyzes can be performed on the accumulated error measures generated by sequences of samples. These collected error measures conventionally provide the Mean Squared Error (MSN) and the Channel Quality Measure (CQM) within the chip.

The error measure output by the slicer is typically accumulated on the read channel chip to evaluate the performance of the chip or memory system. However, in accordance with the present invention, this is done by the slicer 218 generated error measure provided on one of the output pins or output terminals for the chip. This error signal has absolute qualities and is generated at a rate as high as the data rate. As such, the signal is particularly useful in implementing aspects of the present invention.

Referring again to 24 gets the balanced data signal from the FIR equalization filter 216 is output from the data sequence detector 220 output. The illustrated detector is preferably suitable for the data encoding method used in storing the data on the surface of the storage disk. In preferred embodiments of the present invention, a Viterbi decoder for the data sequence detector 220 used. The design and use of such detectors is well known and is illustrated, for example, in the above referenced patent and US Patent No. 5,754,353.

25 illustrates aspects of the decoding operation within a Viterbi decoder. The Trelles-like graph illustrates the time changes of the signal values 222 that in the decoder 220 be entered. From sequences of samples, the decoder determines which values are most likely to have particular sequences of samples. The sequence of most probable values is shown on the graph of 25 with fat points 224 indicated. The adaptation of the most likely data sequence to identify the output data represents the use of a constraint, as that term is used herein and in the despain application discussed above.

The decoder 220 returns the sequence of the best matching data points 224 as the data output from the read channel circuitry. The decoder 220 also generates an error measure for each sample based on the difference between the value assigned to the Viterbi decoder ( 224 ) and the actually received value ( 222 ). This difference 226 is identified in the graph as gray arrows belonging to each of the samples. In normal use, this error measure could 226 be accumulated within the read channel chip as a measure of system performance or channel quality. Conventionally, this measure of error is not available while data is being transferred from the read channel chip.

Preferred embodiments of the read channel of 24 provide the read channel error metric derived from the Viterbi decoder directly to an output port or pin for the read channel chip. As shown in the figure, the slicer error metric may be at an output port 230 are provided, the error measure from the data sequence detector 220 could be from a connection 232 and the output data bit stream could be through one port 234 to be provided. In general, most systems will only be one of the error measurement output signals 230 and 232 lock in. These signals could be provided at an output terminal that is selectable between the two error measurement signals during manufacture. For example, the value passed through an output port may be selectable during manufacture and use of a programmable fuse.

The output error measurement signal is at the terminals 230 . 232 at rates as high as the data rate available and has absolute value characteristics. In fact, that's the measure of error 226 ( 25 ) functionally the error map, which in 13 shown is very similar. As a result, this output becomes 226 over the connections 230 or 232 ( 24 ) particularly useful in the practice of the present invention.

It It should be noted that the error measure, which is an output from provided to the read channel chip, may not be continuously available must, because there are reasons gives why the connection to other uses during a Multiplexing is switched. Suitable error measurement signals are of the preferred read channel chip at a higher data rate as conventional Servo signals available and are available while the data is read and when data is being output from the chip become. additionally it is preferred that the read channel chip process and output from conventional linear servo information continues in the conventional manner, so that they are for use in the manner described above available is when the linear servo information combined with absolute value data is used. If, of course all Servo operations are performed without the linear servo information, would not Necessity, a processing of linear servo information provide within the read channel chip.

Even though the present invention in detail referring to the present preferred embodiments those skilled in the art will be familiar with the art Recognize that various modifications are being made can, without departing from the invention. While the disclosure servo control systems according to the invention with regard to hard disk drive (HDD) applications have systems according to the invention e.g. equally applicable to other servo control applications. Further is the present invention in terms of a hard disk drive in which both the linear servo and absolute value information in Successions of a track can be used. This need must not fulfilled be. For example, you can all operations without using the linear servo operations. It is further conceivable that no conventional servo information is on at least parts of a disk within a disk drive, that embodies aspects of the present invention. As a result, the invention is defined by the following claims and is not to the particularly preferred embodiments described herein limited.

Claims (5)

A control system for use in a controlled variable setting representing a system to be controlled, wherein the controlled variable is represented within an observable variable and at least two values of the controlled variable correspond to a single value of the observable variable, the control system comprising: a signal source ( 126 . 142 ) providing first and second values of the observable variables, the second value of the observable variable temporally after the first value of the observable variable; and a mapping logic ( 148 ), wherein the mapping logic receives the first value of the observable variable and outputs a first and a second possible value of the controlled variable; characterized in that the control system further comprises a first estimator ( 150 ) which is capable of estimating a future state of the system to be controlled, the first estimator taking as an input the first possible value of the controlled variable and a first output variable representing a first predicted value of the observable variable in response to the first generated first possible value of the controlled variable; a second appraiser ( 152 ) which is capable of estimating a future state of the system to be controlled, the second estimator taking as an input the second possible value of the controlled variable and a second output variable representing a second predicted value of the observable variable in response to the second variable second possible value de controlled variables generated; and a determination logic ( 154 ) which determines which of the first and second predicted values of the observable variable corresponds more closely to the second value of the observable variable. A control system according to claim 1, wherein the determination logic ( 154 ) compares an absolute value of the first predicted value with the second value of the observable variable and compares an absolute value of the second predicted value with the second value of the observable variable to determine which of the first and second predicted values of the observable variable is the second value corresponds to the observable variable. The control system of claim 1, wherein the control system maintains the controlled variable close to a nominal value and the nominal value between the first and second possible values of the controlled Variables is. The control system of claim 3, wherein the control system operates in accordance with discrete time intervals, and wherein the first and second estimators ( 150 . 152 ) receives the first and second possible values of the controlled variables only after the control system determines that the controlled variable will assume values above and below the nominal value. A control system according to claim 1, provided within a magnetic data storage system and with a servo system for positioning a magnetic head ( 116 ) relative to a track ( 114 ) which is movable relative to the head, wherein the track ( 114 ) a sequence of bursts of servo signals ( 40 ) along there and data signals ( 46 ) between the bursts of servo signals ( 40 ), comprising the combination of: a device ( 126 ) responsive to the passage of the bursts of servo signals at the magnetic head for generating a first set of error signals; a facility ( 158 ) due to the passing of the data signals on the magnetic head ( 116 ) for generating a second set of error signals; and a facility ( 162 . 140 ), which is responsive to the first and second sets of error signals, for applying the position error signals from the first and second sets to determine the position of the magnetic head ( 116 ) relative to the track ( 114 ) to correct.